New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
C #3620
base: main
Are you sure you want to change the base?
C #3620
Commits on Dec 6, 2022
-
Configuration menu - View commit details
-
Copy full SHA for 0d226e0 - Browse repository at this point
Copy the full SHA 0d226e0View commit details -
Merge pull request open-mmlab#2385 from open-mmlab/dev-1.x
Merge MMSegmentation 1.x development branch dev-1.x to main branch 1.x for v1.0.0rc2
Configuration menu - View commit details
-
Copy full SHA for 4b1eda7 - Browse repository at this point
Copy the full SHA 4b1eda7View commit details
Commits on Dec 30, 2022
-
CodeCamp open-mmlab#144 [Doc] Chinese version of config tutorial (2371)
* [Doc]Translate the 1_config.md and modify a wrong statement in 1_config.md * Translate the 1_config.md and modify a wrong statement in 1_config.md * Modify some expressions * Apply suggestions from code review
Configuration menu - View commit details
-
Copy full SHA for d15e2ae - Browse repository at this point
Copy the full SHA d15e2aeView commit details -
CodeCamp open-mmlab#147 [Doc] Add Chinese version of train & test tut…
…orial (2355) doc modify part of content changed parts of content modified Update docs/zh_cn/user_guides/4_train_test.md Co-authored-by: 谢昕辰 <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for e267350 - Browse repository at this point
Copy the full SHA e267350View commit details -
CodeCamp open-mmlab#1562 [Doc] Add Chinese version of overview.md (op…
…en-mmlab#2397) * CodeCamp open-mmlab#1562 [Doc] update `overview.md` * Update overview.md * Update docs/zh_cn/overview.md Co-authored-by: 谢昕辰 <xiexinch@outlook.com> * Update docs/zh_cn/overview.md Co-authored-by: 谢昕辰 <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 76ce958 - Browse repository at this point
Copy the full SHA 76ce958View commit details -
[CI] Add torch1.13 checking in CI (open-mmlab#2402)
* Add torch1.13 in CI * use mim install mm packages * install all requirements * install wheel * add ref
Configuration menu - View commit details
-
Copy full SHA for f745a43 - Browse repository at this point
Copy the full SHA f745a43View commit details -
[Doc] Add ZN datasets.md in dev-1.x (open-mmlab#2387)
* [Doc] Add ZN datasets.md in dev-1.x * fix typo * fix * fix
Configuration menu - View commit details
-
Copy full SHA for f3a0201 - Browse repository at this point
Copy the full SHA f3a0201View commit details -
CodeCamp open-mmlab#1565 [Doc] update the Chinese version of get_star…
…ted.md (open-mmlab#2417) * DOC Update docs/zh_cn/get_started.md Co-authored-by: 谢昕辰 <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for e3420d7 - Browse repository at this point
Copy the full SHA e3420d7View commit details -
[Projects] Add 'Projects/' folder, and the first example project (ope…
…n-mmlab#2412) add example project add ci ignore add version limits
Configuration menu - View commit details
-
Copy full SHA for 4b81f3b - Browse repository at this point
Copy the full SHA 4b81f3bView commit details -
[Doc] Add dataflow document (open-mmlab#2403)
* draft * update loss * update * add runner * add steps * update
Configuration menu - View commit details
-
Copy full SHA for 4fc4097 - Browse repository at this point
Copy the full SHA 4fc4097View commit details -
[Feature] Add Biomedical 3D array random crop transform (open-mmlab#2378
Configuration menu - View commit details
-
Copy full SHA for 2965e61 - Browse repository at this point
Copy the full SHA 2965e61View commit details -
[Doc] Change some content from customize_runtime to engine ZN doc (op…
…en-mmlab#2341) * [Doc] Change some content from customize_runtime to engine ZN doc * fix comments * add customize runtime setting zn doc * move optimizer content into one section * fix * fix * fix * fix
Configuration menu - View commit details
-
Copy full SHA for a6e3772 - Browse repository at this point
Copy the full SHA a6e3772View commit details -
Configuration menu - View commit details
-
Copy full SHA for 0de3760 - Browse repository at this point
Copy the full SHA 0de3760View commit details -
[Refactor] Support TTA (open-mmlab#2184)
* tta init * use mmcv transform * test city * add multiscale * fix merge * add softmax to post process * add ut * add tta pipeline to other datasets * remove softmax * add encoder_decoder_tta ut * add encoder_decoder_tta ut * rename * rename file * rename config * rm aug_test * move flip to post process * fix channel
Configuration menu - View commit details
-
Copy full SHA for 1aef75b - Browse repository at this point
Copy the full SHA 1aef75bView commit details -
Configuration menu - View commit details
-
Copy full SHA for b7f4fee - Browse repository at this point
Copy the full SHA b7f4feeView commit details -
[Fix] Remove dependcy mmdet when do not use
MaskFormerHead
and `MMD……ET_Mask2FormerHead` (open-mmlab#2448) ## Motivation Calling `mmseg.utils.register_all_modules` will import `MaskFormerHead` and `Mask2FormerHead`, it will crash if mmdet is not installed as `None` cannot be initialized. ## Modification - Modify `MMDET_MaskFormerHead=BaseModule` and `MMDET_Mask2FormerHead = BaseModule` when cannot import from mmdet
Configuration menu - View commit details
-
Copy full SHA for 845df84 - Browse repository at this point
Copy the full SHA 845df84View commit details
Commits on Dec 31, 2022
-
[Fix]Fix pt version for merge stage test (open-mmlab#2449)
## Motivation The conflict is caused by: The user requested torch==1.12.1+cpu torchvision 0.13.0+cpu depends on torch==1.12.0 ## Modification modify the torch==1.12.0+cpu
Configuration menu - View commit details
-
Copy full SHA for bcc5e84 - Browse repository at this point
Copy the full SHA bcc5e84View commit details -
Bump 1.0.0rc3 (open-mmlab#2446)
## Motivation To release 1.0.0rc3 ## Modification 1. Modify mmseg version 2. Add change log 3. Modify README 4. Modify faq 5. Revise docker file
Configuration menu - View commit details
-
Copy full SHA for 6d2f101 - Browse repository at this point
Copy the full SHA 6d2f101View commit details -
Configuration menu - View commit details
-
Copy full SHA for 8354bbc - Browse repository at this point
Copy the full SHA 8354bbcView commit details
Commits on Jan 2, 2023
-
[Feature] nnUNet-style Gaussian Noise and Blur (open-mmlab#2373)
## Motivation implement nnUNet-style Gaussian Noise and Blur
Configuration menu - View commit details
-
Copy full SHA for 85c3c00 - Browse repository at this point
Copy the full SHA 85c3c00View commit details -
[Feature] Add BioMedicalRandomGamma (open-mmlab#2406)
Add the random gamma correction transform for biomedical images, which follows the design of the nnUNet.
Configuration menu - View commit details
-
Copy full SHA for 194f0b9 - Browse repository at this point
Copy the full SHA 194f0b9View commit details
Commits on Jan 3, 2023
-
[Feature] Add BioMedical3DPad (open-mmlab#2383)
## Motivation Add the 3d pad transform for biomedical images, which follows the design of the nnUNet.
Configuration menu - View commit details
-
Copy full SHA for 83ff7d6 - Browse repository at this point
Copy the full SHA 83ff7d6View commit details
Commits on Jan 4, 2023
-
CodeCamp open-mmlab#150 [Feature] Add ISNet (open-mmlab#2400)
## Motivation Support ISNet. paper link: [ISNet: Integrate Image-Level and Semantic-Level Context for Semantic Segmentation](https://openaccess.thecvf.com/content/ICCV2021/papers/Jin_ISNet_Integrate_Image-Level_and_Semantic-Level_Context_for_Semantic_Segmentation_ICCV_2021_paper.pdf) ## Modification Add ISNet decoder head. Add ISNet config.
Configuration menu - View commit details
-
Copy full SHA for 42768f0 - Browse repository at this point
Copy the full SHA 42768f0View commit details
Commits on Jan 6, 2023
-
CodeCamp open-mmlab#140 [New] [Feature] Add synapse dataset and data …
…augmentation in dev-1.x. (open-mmlab#2432) ## Motivation Add Synapse dataset in MMSegmentation. Old PR: open-mmlab#2372.
Configuration menu - View commit details
-
Copy full SHA for 003a088 - Browse repository at this point
Copy the full SHA 003a088View commit details
Commits on Jan 9, 2023
-
[Doc] Add checklist of ISNet (open-mmlab#2460)
## Motivation As title. ## Modification - projects/isnet/README.md
Configuration menu - View commit details
-
Copy full SHA for 0879f2d - Browse repository at this point
Copy the full SHA 0879f2dView commit details
Commits on Jan 10, 2023
-
[CI] Remove test py3.6 (open-mmlab#2468)
## Motivation as title ## Modification 1. .circleci/test.yml 2. .github/workflows/merge_stage_test.yml
Configuration menu - View commit details
-
Copy full SHA for 34444c9 - Browse repository at this point
Copy the full SHA 34444c9View commit details -
[Fix] Fix incorrect
img_shape
value assignment in RandomCrop (open-……mmlab#2469) ## Motivation Fix incorrect `img_shape` value assignment. ## Modification - mmseg/datasets/transforms/transforms.py
Configuration menu - View commit details
-
Copy full SHA for 55e4ae7 - Browse repository at this point
Copy the full SHA 55e4ae7View commit details
Commits on Jan 11, 2023
-
[Doc] Update ZN dataset preparation of Synapse (open-mmlab#2465)
## Motivation - Add Chinese version of Synapse dataset preparation. - Modify all `,` and `。` to `,` and `.` in `docs/zh_cn/user_guides/2_dataset_prepare.md`.
Configuration menu - View commit details
-
Copy full SHA for 9f2d0d2 - Browse repository at this point
Copy the full SHA 9f2d0d2View commit details -
[Feature] Add
gt_edge_map
field. (open-mmlab#2466)## Motivation The motivation of this PR is to add `gt_edge_map` field to support boundary loss. ## Modification - GenerateEdge Modify `gt_edge` field to `gt_edge_map`. - PackSegInputs Add `gt_edge_map` to data_sample. - stack_batch Pad `gt_edge_map` to max_shape. ## BC-breaking (Optional) No ## Use cases (Optional) Reference `GenerateEdge`.
Configuration menu - View commit details
-
Copy full SHA for 6521e36 - Browse repository at this point
Copy the full SHA 6521e36View commit details
Commits on Jan 12, 2023
-
CodeCamp open-mmlab#151[Feature] Support HieraSeg on cityscapes (open…
…-mmlab#2444) ## Support `HieraSeg` interface on `cityscapes` ## Motivation Support `HieraSeg` interface on cityscapes dataset Paper link : https://ieeexplore.ieee.org/document/9878466/ ``` @Article{li2022deep, title={Deep Hierarchical Semantic Segmentation}, author={Li, Liulei and Zhou, Tianfei and Wang, Wenguan and Li, Jianwu and Yang, Yi}, journal={CVPR}, year={2022} } ``` ## Modification Add `HieraSeg_Projects` on `projects/` Add `sep_aspp_contrast_head` decoder head. Add `HieraSeg` config. Add `hiera_loss`, `hiera_triplet_loss_cityscape`, `tree_triplet_loss`
Configuration menu - View commit details
-
Copy full SHA for 933d77a - Browse repository at this point
Copy the full SHA 933d77aView commit details
Commits on Jan 13, 2023
-
[Doc] Fix API document (open-mmlab#2483)
## Motivation As title. ## Modification - docs/en/api.rst - docs/zh_cn/api.rst - add `scipy` to readthedocs requirement.
Configuration menu - View commit details
-
Copy full SHA for c7d1e00 - Browse repository at this point
Copy the full SHA c7d1e00View commit details
Commits on Jan 17, 2023
-
Update basesegdataset.py (open-mmlab#2492)
## Motivation Makes docstring to be consistent with actual argument name. ## Modification Minor fix ## BC-breaking (Optional) No
Configuration menu - View commit details
-
Copy full SHA for d8c18d1 - Browse repository at this point
Copy the full SHA d8c18d1View commit details
Commits on Jan 19, 2023
-
[Fix][Doc] Fix link of preprocessing and order of operations in ZN da…
…taset.md doc. (open-mmlab#2494) ## Motivation Ref: open-mmlab#2464 (comment) Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 42ca580 - Browse repository at this point
Copy the full SHA 42ca580View commit details -
Configuration menu - View commit details
-
Copy full SHA for ec890b2 - Browse repository at this point
Copy the full SHA ec890b2View commit details
Commits on Jan 20, 2023
-
CodeCamp open-mmlab#1555[Feature] Support Mapillary Vistas Dataset (o…
…pen-mmlab#2484) ## Support `Mapillary Vistas Dataset` ## Motivation Support **`Mapillary Vistas Dataset`** Dataset Paper link : https://ieeexplore.ieee.org/document/9878466/ Download and more information view https://www.mapillary.com/dataset/vistas ``` @InProceedings{Neuhold_2017_ICCV, author = {Neuhold, Gerhard and Ollmann, Tobias and Rota Bulo, Samuel and Kontschieder, Peter}, title = {The Mapillary Vistas Dataset for Semantic Understanding of Street Scenes}, booktitle = {Proceedings of the IEEE International Conference on Computer Vision (ICCV)}, month = {Oct}, year = {2017} } ``` ## Modification Add `Mapillary_dataset` in `mmsegmentation/projects` Add `configs/_base_/mapillary_v1_2.py` and `configs/_base_/mapillary_v2_0.py` Add `configs/deeplabv3plus_r18-d8_4xb2-80k_mapillay-512x1024.py` to test training and testing on Mapillary datasets Add `docs/en/user_guides/2_dataset_prepare.md` , add Mapillary Vistas Dataset Preparing and Structure. Add `tools/dataset_converters/mapillary.py` to convert RGB labels to Mask labels. Co-authored-by: 谢昕辰 <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 598a927 - Browse repository at this point
Copy the full SHA 598a927View commit details -
CodeCamp open-mmlab#141 [Feature] Add BioMedical3DRandomFlip. (open-m…
…mlab#2404) ## Motivation Support for biomedical 3d images augmentation. ## Modification Add BioMedical3DRandomFlip in mmseg/datasets/transforms/transforms.py. Co-authored-by: MeowZheng <meowzheng@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 51141d0 - Browse repository at this point
Copy the full SHA 51141d0View commit details -
[Fix] Fix inference api and support setting palette to SegLocalVisual…
…izer (open-mmlab#2475) as title Co-authored-by: MengzhangLI <mcmong@pku.edu.cn>
Configuration menu - View commit details
-
Copy full SHA for 7e0a8c0 - Browse repository at this point
Copy the full SHA 7e0a8c0View commit details
Commits on Jan 22, 2023
-
[Doc] Add EN datasets.md in dev-1.x (open-mmlab#2464)
Add English version of `datasets.md`, the Chinese version is in open-mmlab#2387. Co-authored-by: MeowZheng <meowzheng@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 92f0fbb - Browse repository at this point
Copy the full SHA 92f0fbbView commit details
Commits on Jan 30, 2023
-
[CI] Upgrade the version of isort to fix lint error (open-mmlab#2519)
## Motivation open-mmlab/mmeval#85 --------- Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for b1addef - Browse repository at this point
Copy the full SHA b1addefView commit details -
[Doc] Fix minor typo in migration
package.md
(open-mmlab#2518)Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 2d424b8 - Browse repository at this point
Copy the full SHA 2d424b8View commit details -
[Fix] Switch order of
reduce_zero_label
and applyinglabel_map
in…… 1.x (open-mmlab#2517) This is an almost exact duplicate of open-mmlab#2500 (that was made to the `master` branch) now applied to the `1.x` branch. --- ## Motivation I want to fix a bug through this PR. The bug occurs when two options -- `reduce_zero_label=True`, and custom classes are used. `reduce_zero_label` remaps the GT seg labels by remapping the zero-class to 255 which is ignored. Conceptually, this should occur *before* the `label_map` is applied, which maps *already reduced labels*. However, currently, the `label_map` is applied before the zero label is reduced. ## Modification The modification is simple: - I've just interchanged the order of the two operations by moving a few lines from bottom to top. - I've added a test that passes when the fix is introduced, and fails on the original `master` branch. ## BC-breaking (Optional) I do not anticipate this change braking any backward-compatibility. ## Checklist - [x] Pre-commit or other linting tools are used to fix the potential lint issues. - _I've fixed all linting/pre-commit errors._ - [x] The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. - _I've added a unit test._ - [x] If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. - _I don't think this change affects MMDet or MMDet3D._ - [x] The documentation has been modified accordingly, like docstring or example tutorials. - _This change fixes an existing bug and doesn't require modifying any documentation/docstring._
Configuration menu - View commit details
-
Copy full SHA for a51ddda - Browse repository at this point
Copy the full SHA a51dddaView commit details -
[Fix] Unfinished label conversion from
-1
to255
in 1.x (open-mml……ab#2516) ## Motivation This is motivated by a previously unfinished PR (open-mmlab#2332). In that PR, the label -1 was changed to 255 in `BaseSegDataset`, which is correct. However, it was changed at only one location. There is another location in `mmseg/datasets/basesegdataset.py` where -1 was still being used that was not converted to 255. I have now converted it to 255. This is exactly same as a similar fix to the `master` branch via open-mmlab#2515 . ## Modification I've simply converted the snipped ```python if new_id != -1: new_palette.append(palette[old_id]) ``` to ```python if new_id != 255: new_palette.append(palette[old_id]) ``` ## Checklist - [x] Pre-commit or other linting tools are used to fix the potential lint issues. - _I've fixed all linting/pre-commit errors._ - [x] The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. - _No unit tests need to be added or were affected. - [x] If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. - _I don't think this change affects MMDet or MMDet3D._ - [x] The documentation has been modified accordingly, like docstring or example tutorials. - _This change fixes an existing bug and doesn't require modifying any documentation/docstring._
Configuration menu - View commit details
-
Copy full SHA for bf3a3a1 - Browse repository at this point
Copy the full SHA bf3a3a1View commit details -
Configuration menu - View commit details
-
Copy full SHA for d20305c - Browse repository at this point
Copy the full SHA d20305cView commit details -
Configuration menu - View commit details
-
Copy full SHA for 14710ac - Browse repository at this point
Copy the full SHA 14710acView commit details -
Configuration menu - View commit details
-
Copy full SHA for ee50cd2 - Browse repository at this point
Copy the full SHA ee50cd2View commit details
Commits on Jan 31, 2023
-
Configuration menu - View commit details
-
Copy full SHA for e5c5602 - Browse repository at this point
Copy the full SHA e5c5602View commit details
Commits on Feb 1, 2023
-
Backward merge open-mmlab#2535 from
1.x
intodev-1.x
(open-mmlab#……2546) ## Motivation This is essentially open-mmlab#2535 that I had intended to submit to the `dev-1.x` branch but accidentally submitted it directly to the `1.x` branch (apologies!). This also got approved possibly because the core devs also didn't realize this. The problem is that now `1.x` and `dev-1.x` are out of sync -- the changes introduced by open-mmlab#2535 will never be reflected in `dev-1.x`. ## Modification I'm proposing this "backward-merge" so that `1.x` and `dev-1.x` can be in sync again. If you look at "files changed", they are exactly the changes introduced by open-mmlab#2535. Co-authored-by: MeowZheng <meowzheng@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 105726b - Browse repository at this point
Copy the full SHA 105726bView commit details -
[Refactor] Refactor fileio (open-mmlab#2543)
## Motivation Use the new fileio from mmengine open-mmlab/mmengine#533 ## Modification 1. Use `mmengine.fileio` to repalce FileClient in mmseg/datasets 2. Use `mmengine.fileio` to repalce FileClient in mmseg/datasets/transforms 3. Use `mmengine.fileio` to repalce FileClient in mmseg/visualization ## BC-breaking (Optional) we modify all the dataset configurations, so please use the latest config file.
Configuration menu - View commit details
-
Copy full SHA for 1bad52a - Browse repository at this point
Copy the full SHA 1bad52aView commit details -
[Fix] Fix MaskFormer and Mask2Former of MMSegmentation (open-mmlab#2532)
## Motivation The DETR-related modules have been refactored in open-mmlab/mmdetection#8763, which causes breakings of MaskFormer and Mask2Former in both MMDetection (has been fixed in open-mmlab/mmdetection#9515) and MMSegmentation. This pr fix the bugs in MMSegmentation. ### TO-DO List - [x] update configs - [x] check and modify data flow - [x] fix unit test - [x] aligning inference - [x] write a ckpt converter - [x] write ckpt update script - [x] update model zoo - [x] update model link in readme - [x] update [faq.md](https://github.com/open-mmlab/mmsegmentation/blob/dev-1.x/docs/en/notes/faq.md#installation) ## Tips of Fixing other implementations based on MaskXFormer of mmseg 1. The Transformer modules should be built directly. The original building with register manner has been refactored. 2. The config requires to be modified. Delete `type` and modify several keys, according to the modifications in this pr. 3. The `batch_first` is set `True` uniformly in the new implementations. Hence the data flow requires to be transposed and config of `batch_first` needs to be modified. 4. The checkpoint trained on the old implementation should be converted to be used in the new one. ### Convert script ```Python import argparse from copy import deepcopy from collections import OrderedDict import torch from mmengine.config import Config from mmseg.models import build_segmentor from mmseg.utils import register_all_modules register_all_modules(init_default_scope=True) def parse_args(): parser = argparse.ArgumentParser( description='MMSeg convert MaskXFormer model, by Li-Qingyun') parser.add_argument('Mask_what_former', type=int, help='Mask what former, can be a `1` or `2`', choices=[1, 2]) parser.add_argument('CFG_FILE', help='config file path') parser.add_argument('OLD_CKPT_FILEPATH', help='old ckpt file path') parser.add_argument('NEW_CKPT_FILEPATH', help='new ckpt file path') args = parser.parse_args() return args args = parse_args() def get_new_name(old_name: str): new_name = old_name if 'encoder.layers' in new_name: new_name = new_name.replace('attentions.0', 'self_attn') new_name = new_name.replace('ffns.0', 'ffn') if 'decoder.layers' in new_name: if args.Mask_what_former == 2: # for Mask2Former new_name = new_name.replace('attentions.0', 'cross_attn') new_name = new_name.replace('attentions.1', 'self_attn') else: # for Mask2Former new_name = new_name.replace('attentions.0', 'self_attn') new_name = new_name.replace('attentions.1', 'cross_attn') return new_name def cvt_sd(old_sd: OrderedDict): new_sd = OrderedDict() for name, param in old_sd.items(): new_name = get_new_name(name) assert new_name not in new_sd new_sd[new_name] = param assert len(new_sd) == len(old_sd) return new_sd if __name__ == '__main__': cfg = Config.fromfile(args.CFG_FILE) model_cfg = cfg.model segmentor = build_segmentor(model_cfg) refer_sd = segmentor.state_dict() old_ckpt = torch.load(args.OLD_CKPT_FILEPATH) old_sd = old_ckpt['state_dict'] new_sd = cvt_sd(old_sd) print(segmentor.load_state_dict(new_sd)) new_ckpt = deepcopy(old_ckpt) new_ckpt['state_dict'] = new_sd torch.save(new_ckpt, args.NEW_CKPT_FILEPATH) print(f'{args.NEW_CKPT_FILEPATH} has been saved!') ``` Usage: ```bash # for example python ckpt4pr2532.py 1 configs/maskformer/maskformer_r50-d32_8xb2-160k_ade20k-512x512.py original_ckpts/maskformer_r50-d32_8xb2-160k_ade20k-512x512_20221030_182724-cbd39cc1.pth cvt_outputs/maskformer_r50-d32_8xb2-160k_ade20k-512x512_20221030_182724.pth python ckpt4pr2532.py 2 configs/mask2former/mask2former_r50_8xb2-160k_ade20k-512x512.py original_ckpts/mask2former_r50_8xb2-160k_ade20k-512x512_20221204_000055-4c62652d.pth cvt_outputs/mask2former_r50_8xb2-160k_ade20k-512x512_20221204_000055.pth ``` --------- Co-authored-by: MeowZheng <meowzheng@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for ef981d4 - Browse repository at this point
Copy the full SHA ef981d4View commit details -
Rename typing.py to typing_utils.py (open-mmlab#2548)
## Motivation Fix the bug in running ```collect_evn.py```. open-mmlab#2547 ## Modification Rename typing.py to typing_utils.py
Configuration menu - View commit details
-
Copy full SHA for d9dfa27 - Browse repository at this point
Copy the full SHA d9dfa27View commit details -
Configuration menu - View commit details
-
Copy full SHA for a617033 - Browse repository at this point
Copy the full SHA a617033View commit details -
Configuration menu - View commit details
-
Copy full SHA for 1f108e9 - Browse repository at this point
Copy the full SHA 1f108e9View commit details
Commits on Feb 3, 2023
-
CodeCamp open-mmlab#139 [Feature] Support REFUGE dataset. (open-mmlab…
…#2554) ## Motivation Add REFUGE datasets Old PR: open-mmlab#2420 --------- Co-authored-by: MengzhangLI <mcmong@pku.edu.cn>
Configuration menu - View commit details
-
Copy full SHA for 674b0f0 - Browse repository at this point
Copy the full SHA 674b0f0View commit details
Commits on Feb 6, 2023
-
[Fix] Rename and Fix bug of projects HieraSeg (old PR open-mmlab#2444) (
open-mmlab#2565) ## Motivation Supplementary PR open-mmlab#2444 Fix tiny bug and add loss_by_feat() to compute loss to train. The inference process have verified to be accurate. ## Modification - modify `sep_aspp_contrast_head.py` , add `loss_by_feat()` function to train(training still has bug, will fix in future😫) - fix testing commands path error `bash tools/dist_test.sh projects/HieraSeg_project/` to `bash tools/dist_test.sh projects/HieraSeg/` at README.md
Configuration menu - View commit details
-
Copy full SHA for 1a010e3 - Browse repository at this point
Copy the full SHA 1a010e3View commit details
Commits on Feb 7, 2023
-
[Doc] Add EN custmized runtime doc in dev-1.x (open-mmlab#2533)
## Motivation Translate Chinese version customized runtime doc into English open-mmlab#2502.
Configuration menu - View commit details
-
Copy full SHA for e52ffa2 - Browse repository at this point
Copy the full SHA e52ffa2View commit details -
[NPU] add npu result (open-mmlab#2569)
Motivation add NPU results. Modification add docs/en/device/npu.md and docs/zh_cn/device/npu.md that accompanies the submission results. --------- Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 808ce5b - Browse repository at this point
Copy the full SHA 808ce5bView commit details
Commits on Feb 8, 2023
-
[Doc] for Visualization feature map using wandb backend in dev-1.x (o…
…pen-mmlab#2557) ## Motivation Docs for Visualization featusre map using wandb backend. ## Modification Add a new markdown file and result demo of wandb. --------- Co-authored-by: MeowZheng <meowzheng@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 4504bc7 - Browse repository at this point
Copy the full SHA 4504bc7View commit details
Commits on Feb 10, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 8ec0c25 - Browse repository at this point
Copy the full SHA 8ec0c25View commit details -
Configuration menu - View commit details
-
Copy full SHA for 6f1dab1 - Browse repository at this point
Copy the full SHA 6f1dab1View commit details
Commits on Feb 15, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 382f9b0 - Browse repository at this point
Copy the full SHA 382f9b0View commit details -
Configuration menu - View commit details
-
Copy full SHA for 7e94dda - Browse repository at this point
Copy the full SHA 7e94ddaView commit details -
[NPU] add npu result (open-mmlab#2596)
Motivation add NPU results -- apcnet, bisenetv1, bisenetv2 Modification add docs/en/device/npu.md and docs/zh_cn/device/npu.md that accompanies the submission results. --------- Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 6954192 - Browse repository at this point
Copy the full SHA 6954192View commit details -
[Enhancement]Replace numpy ascontiguousarray with torch contiguous to…
… speed-up (open-mmlab#2604) ## Motivation Original motivation was after [MMDetection PR #9533](open-mmlab/mmdetection#9533) With several experiments I found out that if a ndarray is contiguous, numpy.transpose + torch.contiguous perform better, while if not, then use numpy.ascontiguousarray + numpy.transpose ## Modification Replace numpy.ascontiguousarray with torch.contiguous in [PackSegInputs](https://github.com/open-mmlab/mmsegmentation/blob/1.x/mmseg/datasets/transforms/formatting.py) Co-authored-by: MeowZheng <meowzheng@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 52a2807 - Browse repository at this point
Copy the full SHA 52a2807View commit details
Commits on Feb 16, 2023
-
[FIx] Set default
backend_args
values to None (open-mmlab#2597)## Motivation In MMEngine >= 0.2.0, it might directly determine what the backend is by using the `data_root` path. ## Modification Set all default `backend_args` values are `None`.
Configuration menu - View commit details
-
Copy full SHA for 3a0e7be - Browse repository at this point
Copy the full SHA 3a0e7beView commit details -
[DOC] update link in NPU DOC (open-mmlab#2610)
## Motivation update link in dock ## Modification docs/en/device/npu.md docs/zh_cn/device/npu.md
Configuration menu - View commit details
-
Copy full SHA for 751d307 - Browse repository at this point
Copy the full SHA 751d307View commit details
Commits on Feb 22, 2023
-
[Enhancement] Refine projects (open-mmlab#2586)
## Motivation Make projects contribution more clear ## Modification 1. Add description on project/README 2. Modify comments to reference in example_project/README 3. Add faq for projects ## BC-breaking (Optional) No
Configuration menu - View commit details
-
Copy full SHA for 5750e4d - Browse repository at this point
Copy the full SHA 5750e4dView commit details
Commits on Feb 23, 2023
-
[Feature] Support auto import modules from registry. (open-mmlab#2481)
## Motivation The registry now supports auto-import modules from the given location. register_all_modules before running is no longer needed. The modules will be lazy-imported during building. - [x] This PR can be merged after open-mmlab/mmengine#643. The MMEngine version should be updated. Ref: open-mmlab/mmdetection#9143
Configuration menu - View commit details
-
Copy full SHA for c746a86 - Browse repository at this point
Copy the full SHA c746a86View commit details -
[Feature] Support MMSegInferencer (open-mmlab#2413)
## Motivation Support `MMSegInferencer` for providing an easy and clean interface for single or multiple images inferencing. Ref: open-mmlab/mmengine#773 open-mmlab/mmocr#1608 ## Modification - mmseg/apis/mmseg_inferencer.py - mmseg/visualization/local_visualizer.py - demo/image_demo_with_inferencer.py ## Use cases (Optional) Based on https://github.com/open-mmlab/mmengine/tree/inference Add a new image inference demo with `MMSegInferencer` - demo/image_demo_with_inferencer.py ```shell python demo/image_demo_with_inferencer.py demo/demo.png fcn_r50-d8_4xb2-40k_cityscapes-512x1024 ``` --------- Co-authored-by: MeowZheng <meowzheng@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for e75b8ac - Browse repository at this point
Copy the full SHA e75b8acView commit details
Commits on Feb 24, 2023
-
[Enhancement] Remove mmdet and mmcls from mminstall (open-mmlab#2642)
## Motivation As the mmdet and mmcls are not very stabel, and mim can install repo from source code, we remove them from mminstall and they won't be installed automatically when run `mim install mmsegmentation` ## Modification 1. remove mmdet and mcls from mminstall 2. add explanation in faq --------- Co-authored-by: MengzhangLI <mcmong@pku.edu.cn>
Configuration menu - View commit details
-
Copy full SHA for d4cda68 - Browse repository at this point
Copy the full SHA d4cda68View commit details
Commits on Feb 27, 2023
-
tools/analysis_tools browse_dataset.py (open-mmlab#2649)
## Motivation browse_dataset before training ## Modification create tools/analysis_tools/browse_dataset.py
Configuration menu - View commit details
-
Copy full SHA for da98d8d - Browse repository at this point
Copy the full SHA da98d8dView commit details
Commits on Feb 28, 2023
-
[Fix] Add out_channels in
CascadeEncoderDecoder
and update OCRNet a……nd MobileNet v2 results (open-mmlab#2656) ## Motivation As title. ## Modification 1. update results in readme 2. fix attr error in cascade encoder decoder
Configuration menu - View commit details
-
Copy full SHA for d7dd1e2 - Browse repository at this point
Copy the full SHA d7dd1e2View commit details
Commits on Mar 2, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 760d0bf - Browse repository at this point
Copy the full SHA 760d0bfView commit details -
[Docs] Add Chinese dataflow markdown (open-mmlab#2652)
## Modification Add Chinese dataflow markdown --------- Signed-off-by: csatsurnh <cshan1995@126.com> Co-authored-by: csatsurnh <cshan1995@126.com> Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 5d16af3 - Browse repository at this point
Copy the full SHA 5d16af3View commit details
Commits on Mar 3, 2023
-
[Enhancement] Modify interface of MMSeginferencer and add docs (open-…
…mmlab#2658) ## Motivation Make MMSeginferencer easier to be used ## Modification 1. Add `_load_weights_to_model` to MMSeginferencer, it is for get `dataset_meta` from ckpt 2. Modify and remove some parameters of `__call__`, `visualization` and `postprocess` 3. Add function of save seg mask, remove dump pkl. 4. Refine docstring of MMSeginferencer and SegLocalVisualizer 5. Add the user documentation of MMSeginferencer ## BC-breaking (Optional) yes, remove some parameters, we need to discuss whether keep them with deprecated waring or just remove them as the MMSeginferencer just merged in mmseg a few days. Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 53c9c09 - Browse repository at this point
Copy the full SHA 53c9c09View commit details -
Configuration menu - View commit details
-
Copy full SHA for 7872468 - Browse repository at this point
Copy the full SHA 7872468View commit details -
Configuration menu - View commit details
-
Copy full SHA for 616dc86 - Browse repository at this point
Copy the full SHA 616dc86View commit details -
Configuration menu - View commit details
-
Copy full SHA for e134950 - Browse repository at this point
Copy the full SHA e134950View commit details
Commits on Mar 6, 2023
-
Configuration menu - View commit details
-
Copy full SHA for ef0d712 - Browse repository at this point
Copy the full SHA ef0d712View commit details -
Configuration menu - View commit details
-
Copy full SHA for 695c78b - Browse repository at this point
Copy the full SHA 695c78bView commit details
Commits on Mar 7, 2023
-
[Doc] Add zh_cn models doc and fix en doc typo (open-mmlab#2703)
as title --------- Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 416aa59 - Browse repository at this point
Copy the full SHA 416aa59View commit details -
[Fix] Support format_result and fix prefix param in cityscape metric,…
… and rename CitysMetric to CityscapesMetric (open-mmlab#2660) as title
Configuration menu - View commit details
-
Copy full SHA for f45417f - Browse repository at this point
Copy the full SHA f45417fView commit details -
Configuration menu - View commit details
-
Copy full SHA for faf0833 - Browse repository at this point
Copy the full SHA faf0833View commit details -
en doc of uisualization_feature_map.md (open-mmlab#2715)
## Motivation En doc for visualization_feature_map.md and index.rst ## Modification Add new file and change index.rst
Configuration menu - View commit details
-
Copy full SHA for f56ddca - Browse repository at this point
Copy the full SHA f56ddcaView commit details
Commits on Mar 9, 2023
-
[Doc] Add zh_cn evaluation doc and fix en typo (open-mmlab#2701)
as title --------- Signed-off-by: csatsurnh <cshan1995@126.com>
Configuration menu - View commit details
-
Copy full SHA for a5bcd85 - Browse repository at this point
Copy the full SHA a5bcd85View commit details -
Configuration menu - View commit details
-
Copy full SHA for a1fa656 - Browse repository at this point
Copy the full SHA a1fa656View commit details
Commits on Mar 10, 2023
-
[Feature] Support calculating FLOPs of segmentors (open-mmlab#2706)
## Motivation fix compute flops problems ## Modification Please briefly describe what modification is made in this PR.
Configuration menu - View commit details
-
Copy full SHA for 45e79a0 - Browse repository at this point
Copy the full SHA 45e79a0View commit details
Commits on Mar 13, 2023
-
[Doc] add zh_cn migration doc (open-mmlab#2733)
as title --------- Co-authored-by: MeowZheng <meowzheng@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 2cbccac - Browse repository at this point
Copy the full SHA 2cbccacView commit details
Commits on Mar 14, 2023
-
[Enhancement] Support input gt seg map is not 2D (open-mmlab#2739)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation fix open-mmlab#2593 ## Modification 1. Only when gt seg map is 2D, extend its shape to 3D PixelData 2. If seg map is not 2D, we raised warning for users. --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 344fa05 - Browse repository at this point
Copy the full SHA 344fa05View commit details -
[Typo] Change indexes to indices (open-mmlab#2747)
## Modification I just replaced the `indexes` variable name with `indices` for naming consistency.
Configuration menu - View commit details
-
Copy full SHA for c816643 - Browse repository at this point
Copy the full SHA c816643View commit details
Commits on Mar 15, 2023
-
[Datasets] Add Mapillary Vistas Datasets to MMSeg Core Package. (open…
…-mmlab#2576) ## [Datasets] Add Mapillary Vistas Datasets to MMSeg Core Package . ## Motivation Add Mapillary Vistas Datasets to core package. Old PR open-mmlab#2484 ## Modification - Add Mapillary Vistas Datasets to core package. - Delete `tools/datasets_convert/mapillary.py` , dataset does't need converting. - Add `schedule_240k.py` config. - Add configs files. ```none deeplabv3plus_r101-d8_4xb2-240k_mapillay_v1-512x1024.py deeplabv3plus_r101-d8_4xb2-240k_mapillay_v2-512x1024.py maskformer_swin-s_4xb2-240k_mapillary_v1-512x1024.py maskformer_swin-s_4xb2-240k_mapillary_v2-512x1024.py maskformer_r101-d8_4xb2-240k_mapillary_v1-512x1024.py maskformer_r101-d8_4xb2-240k_mapillary_v2-512x1024.py pspnet_r101-d8_4xb2-240k_mapillay_v1-512x1024.py pspnet_r101-d8_4xb2-240k_mapillay_v2-512x1024.py ``` - Synchronized changes to `projects/mapillary_datasets` --------- Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com> Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for a39f43e - Browse repository at this point
Copy the full SHA a39f43eView commit details -
[Feature] Support PIDNet (open-mmlab#2609)
## Motivation Support SOTA real-time semantic segmentation method in [Paper with code](https://paperswithcode.com/task/real-time-semantic-segmentation) Paper: https://arxiv.org/pdf/2206.02066.pdf Official repo: https://github.com/XuJiacong/PIDNet ## Current results **Cityscapes** |Model|Ref mIoU|mIoU (ours)| |---|---|---| |PIDNet-S|78.8|78.74| |PIDNet-M|79.9|80.22| |PIDNet-L|80.9|80.89| ## TODO - [x] Support inference with official weights - [x] Support training on Cityscapes - [x] Update docstring - [x] Add unit test
Configuration menu - View commit details
-
Copy full SHA for 4756fcf - Browse repository at this point
Copy the full SHA 4756fcfView commit details -
Configuration menu - View commit details
-
Copy full SHA for 90a7704 - Browse repository at this point
Copy the full SHA 90a7704View commit details -
[Dev] Replace the test images (open-mmlab#2754)
## Motivation The original images are too large. ## Modification Crop to small images.
Configuration menu - View commit details
-
Copy full SHA for 1f654b7 - Browse repository at this point
Copy the full SHA 1f654b7View commit details -
[Enhance] Support multi-band image for Mosaic (open-mmlab#2748)
## Modification I changed the hardcoded 3 channel length to dynamic channel length in `np.full` function arguments. This modification enables `RandomMosaic` transform to support multispectral image (e.g. RGB image with NIR band) or bi-temporal image pairs for change detection task. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for 2c12fdf - Browse repository at this point
Copy the full SHA 2c12fdfView commit details
Commits on Mar 16, 2023
-
[Feature] Support SegNeXt in MMSegmentation 2.0 (open-mmlab#2654)
## Motivation Support SegNeXt in MMSeg 1.x branch. 0.x PR: open-mmlab#2600 --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for e1b15ad - Browse repository at this point
Copy the full SHA e1b15adView commit details
Commits on Mar 17, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 0dc8202 - Browse repository at this point
Copy the full SHA 0dc8202View commit details -
Configuration menu - View commit details
-
Copy full SHA for 861908d - Browse repository at this point
Copy the full SHA 861908dView commit details -
[Features]Support dump segment predition (open-mmlab#2712)
## Motivation 1. It is used to save the segmentation predictions as files and upload these files to a test server ## Modification 1. Add output_file and format only in `IoUMetric` ## BC-breaking (Optional) No ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 3. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 4. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 5. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for 19d1b3e - Browse repository at this point
Copy the full SHA 19d1b3eView commit details
Commits on Mar 21, 2023
-
Configuration menu - View commit details
-
Copy full SHA for ee7c4dc - Browse repository at this point
Copy the full SHA ee7c4dcView commit details
Commits on Mar 22, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 44fcf82 - Browse repository at this point
Copy the full SHA 44fcf82View commit details
Commits on Mar 23, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 121c971 - Browse repository at this point
Copy the full SHA 121c971View commit details
Commits on Mar 30, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 1035a3a - Browse repository at this point
Copy the full SHA 1035a3aView commit details -
[Doc] Fix invalid links and standardize Chinese and English punctuati…
…on marks at zh_CN faq.md (open-mmlab#2790)
Configuration menu - View commit details
-
Copy full SHA for e82b4cb - Browse repository at this point
Copy the full SHA e82b4cbView commit details -
Configuration menu - View commit details
-
Copy full SHA for e96a7d6 - Browse repository at this point
Copy the full SHA e96a7d6View commit details -
Configuration menu - View commit details
-
Copy full SHA for d8f188a - Browse repository at this point
Copy the full SHA d8f188aView commit details -
Configuration menu - View commit details
-
Copy full SHA for c63f5b7 - Browse repository at this point
Copy the full SHA c63f5b7View commit details -
Configuration menu - View commit details
-
Copy full SHA for ff9662c - Browse repository at this point
Copy the full SHA ff9662cView commit details -
Configuration menu - View commit details
-
Copy full SHA for 0689270 - Browse repository at this point
Copy the full SHA 0689270View commit details
Commits on Mar 31, 2023
-
[Doc] Refine doc and fix links (open-mmlab#2821)
## Motivation - Create the `main` branch ## Modification Modify links from `dev-1.x` to `main`
Configuration menu - View commit details
-
Copy full SHA for a1a2027 - Browse repository at this point
Copy the full SHA a1a2027View commit details -
Configuration menu - View commit details
-
Copy full SHA for ae35942 - Browse repository at this point
Copy the full SHA ae35942View commit details
Commits on Apr 3, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 7c8ee79 - Browse repository at this point
Copy the full SHA 7c8ee79View commit details -
Configuration menu - View commit details
-
Copy full SHA for 79635e4 - Browse repository at this point
Copy the full SHA 79635e4View commit details -
[Fix] Remove locations of not exists modules in the registry (open-mm…
…lab#2829) ## Motivation If the module does not actually exist, setting locations will report an error. open-mmlab/mmengine#1010 ## Modification mmseg/registry/registry.py
Configuration menu - View commit details
-
Copy full SHA for 2a657df - Browse repository at this point
Copy the full SHA 2a657dfView commit details -
Configuration menu - View commit details
-
Copy full SHA for 8b20a03 - Browse repository at this point
Copy the full SHA 8b20a03View commit details -
Configuration menu - View commit details
-
Copy full SHA for 4fcb7aa - Browse repository at this point
Copy the full SHA 4fcb7aaView commit details
Commits on Apr 6, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 17e665d - Browse repository at this point
Copy the full SHA 17e665dView commit details -
[Doc] Update readme (open-mmlab#2834)
## Motivation As title, lead users to follow our migration document. ## Checklist - [x] open-mmlab#2801
Configuration menu - View commit details
-
Copy full SHA for 6715b04 - Browse repository at this point
Copy the full SHA 6715b04View commit details -
Configuration menu - View commit details
-
Copy full SHA for bab91bf - Browse repository at this point
Copy the full SHA bab91bfView commit details -
Configuration menu - View commit details
-
Copy full SHA for c72945f - Browse repository at this point
Copy the full SHA c72945fView commit details -
Configuration menu - View commit details
-
Copy full SHA for 7c8af0a - Browse repository at this point
Copy the full SHA 7c8af0aView commit details -
Configuration menu - View commit details
-
Copy full SHA for 6c6b016 - Browse repository at this point
Copy the full SHA 6c6b016View commit details
Commits on Apr 13, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 1c7d223 - Browse repository at this point
Copy the full SHA 1c7d223View commit details -
Configuration menu - View commit details
-
Copy full SHA for 42207bd - Browse repository at this point
Copy the full SHA 42207bdView commit details -
Configuration menu - View commit details
-
Copy full SHA for b7668c1 - Browse repository at this point
Copy the full SHA b7668c1View commit details
Commits on Apr 14, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 116c9c2 - Browse repository at this point
Copy the full SHA 116c9c2View commit details
Commits on Apr 18, 2023
-
Configuration menu - View commit details
-
Copy full SHA for d0ff7c2 - Browse repository at this point
Copy the full SHA d0ff7c2View commit details -
Configuration menu - View commit details
-
Copy full SHA for a9af1d0 - Browse repository at this point
Copy the full SHA a9af1d0View commit details
Commits on Apr 19, 2023
-
Configuration menu - View commit details
-
Copy full SHA for a76b12d - Browse repository at this point
Copy the full SHA a76b12dView commit details -
Configuration menu - View commit details
-
Copy full SHA for e04b841 - Browse repository at this point
Copy the full SHA e04b841View commit details -
Configuration menu - View commit details
-
Copy full SHA for 98b19ea - Browse repository at this point
Copy the full SHA 98b19eaView commit details -
Configuration menu - View commit details
-
Copy full SHA for a8e62f3 - Browse repository at this point
Copy the full SHA a8e62f3View commit details -
Configuration menu - View commit details
-
Copy full SHA for 8327e29 - Browse repository at this point
Copy the full SHA 8327e29View commit details
Commits on Apr 23, 2023
-
Configuration menu - View commit details
-
Copy full SHA for da09304 - Browse repository at this point
Copy the full SHA da09304View commit details -
[Fix] fix squeeze error when N=1 and C=1 (open-mmlab#2933)
## Motivation fix squeeze error when N=1 and C=1 ## Modification fix squeeze error when N=1 and C=1
Configuration menu - View commit details
-
Copy full SHA for 8c8877d - Browse repository at this point
Copy the full SHA 8c8877dView commit details
Commits on Apr 24, 2023
-
[Feature] Support albu transform (open-mmlab#2943)
## Motivation https://github.com/open-mmlab/mmsegmentation/blob/3a14c3597418d3e6eb64a3c65044fcf91f0c105d/mmseg/datasets/pipelines/transforms.py#L1348 ## Modification Add albu to dev-1.x
Configuration menu - View commit details
-
Copy full SHA for ce1e3f7 - Browse repository at this point
Copy the full SHA ce1e3f7View commit details
Commits on Apr 26, 2023
-
[Doc] update repo list in README (open-mmlab#2949)
## Motivation Update repo information and URLs in README. ## Modification ## BC-breaking (Optional) ## Use cases (Optional) --------- Co-authored-by: CSH <40987381+csatsurnh@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 25d814c - Browse repository at this point
Copy the full SHA 25d814cView commit details -
Fix typo in docs/en/user_guides/visualization_feature_map.md (open-mm…
…lab#2951) Motivation Typo in docs/en/user_guides/visualization_feature_map.md. Modification reature -> feature Checklist - [x] Pre-commit or other linting tools are used to fix the potential lint issues. - [x] The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. - [x] If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. - [x] The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for 820c555 - Browse repository at this point
Copy the full SHA 820c555View commit details
Commits on Apr 27, 2023
-
[Feature] Support DDRNet (open-mmlab#2855)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Support DDRNet Paper: [Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes](https://arxiv.org/pdf/2101.06085) official Code: https://github.com/ydhongHIT/DDRNet There is already a PR open-mmlab#1722 , but it has been inactive for a long time. ## Current Result ### Cityscapes #### inference with converted official weights | Method | Backbone | mIoU(official) | mIoU(converted weight) | | ------ | ------------- | -------------- | ---------------------- | | DDRNet | DDRNet23-slim | 77.8 | 77.84 | | DDRNet | DDRNet23 | 79.5 | 79.53 | #### training with converted pretrained backbone | Method | Backbone | Crop Size | Lr schd | Inf time(fps) | Device | mIoU | mIoU(ms+flip) | config | download | | ------ | ------------- | --------- | ------- | ------- | -------- | ----- | ------------- | ------------ | ------------ | | DDRNet | DDRNet23-slim | 1024x1024 | 120000 | 85.85 | RTX 8000 | 77.85 | 79.80 | [config](https://github.com/whu-pzhang/mmsegmentation/blob/ddrnet/configs/ddrnet/ddrnet_23-slim_in1k-pre_2xb6-120k_cityscapes-1024x1024.py) | model \| log | | DDRNet | DDRNet23 | 1024x1024 | 120000 | 33.41 | RTX 8000 | 79.53 | 80.98 | [config](https://github.com/whu-pzhang/mmsegmentation/blob/ddrnet/configs/ddrnet/ddrnet_23_in1k-pre_2xb6-120k_cityscapes-1024x1024.py) | model \| log | The converted pretrained backbone weights download link: 1. [ddrnet23s_in1k_mmseg.pth](https://drive.google.com/file/d/1Ni4F1PMGGjuld-1S9fzDTmneLfpMuPTG/view?usp=sharing) 2. [ddrnet23_in1k_mmseg.pth](https://drive.google.com/file/d/11rsijC1xOWB6B0LgNQkAG-W6e1OdbCyJ/view?usp=sharing) ## To do - [x] support inference with converted official weights - [x] support training on cityscapes dataset --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 3bc5cc4 - Browse repository at this point
Copy the full SHA 3bc5cc4View commit details
Commits on May 4, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 82f48d9 - Browse repository at this point
Copy the full SHA 82f48d9View commit details
Commits on May 5, 2023
-
Configuration menu - View commit details
-
Copy full SHA for a968b5c - Browse repository at this point
Copy the full SHA a968b5cView commit details -
Configuration menu - View commit details
-
Copy full SHA for 869c6ee - Browse repository at this point
Copy the full SHA 869c6eeView commit details
Commits on May 6, 2023
-
Configuration menu - View commit details
-
Copy full SHA for b40efb8 - Browse repository at this point
Copy the full SHA b40efb8View commit details -
Configuration menu - View commit details
-
Copy full SHA for f55cc2f - Browse repository at this point
Copy the full SHA f55cc2fView commit details -
[Fix] Fix DDRNet readme (open-mmlab#2981)
Ref: https://github.com/ydhongHIT/DDRNet#citation Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for cef3255 - Browse repository at this point
Copy the full SHA cef3255View commit details -
Configuration menu - View commit details
-
Copy full SHA for 95d1949 - Browse repository at this point
Copy the full SHA 95d1949View commit details
Commits on May 8, 2023
-
[Feature] Add GDAL backend and Support LEVIR-CD Dataset (open-mmlab#2903
) ## Motivation For support with reading multiple remote sensing image formats, please refer to https://gdal.org/drivers/raster/index.html. Byte, UInt16, Int16, UInt32, Int32, Float32, Float64, CInt16, CInt32, CFloat32 and CFloat64 are supported for reading and writing. Support input of two images for change detection tasks, and support the LEVIR-CD dataset. ## Modification Add LoadSingleRSImageFromFile in 'mmseg/datasets/transforms/loading.py'. Load a single remote sensing image for object segmentation tasks. Add LoadMultipleRSImageFromFile in 'mmseg/datasets/transforms/loading.py'. Load two remote sensing images for change detection tasks. Add ConcatCDInput in 'mmseg/datasets/transforms/transforms.py'. Combine images that have been separately augmented for data enhancement. Add BaseCDDataset in 'mmseg/datasets/basesegdataset.py' Base class for datasets used in change detection tasks. --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 71fa204 - Browse repository at this point
Copy the full SHA 71fa204View commit details
Commits on May 10, 2023
-
[Feature] Support DSDL Dataset (open-mmlab#2925)
- support dsdl seg dataset - add dsdl dataset citest - validated accuracy on voc2012 and cityscapes
Configuration menu - View commit details
-
Copy full SHA for facba6d - Browse repository at this point
Copy the full SHA facba6dView commit details
Commits on May 12, 2023
-
[Feature] Prevent passed
ann_file
from silently failing to load (op……en-mmlab#2966) ## Motivation While customizing the number of samples using `ann_file` for Cityscapes, I noticed that when the `ann_file` name is incorrect, it will silently resort to loading the dataset from the directory. I think when the user intends to load using `ann_file`, it should not silently fail, but give some sort of error message or warning. ## Modification I added assertion to check whether the `ann_file` exists instead of silently resorting to loading from the directory. Since `ann_file` is set to `''` by default and joined with `self.data_root`, I used `osp.isdir` to first check if `self.ann_dir` is a directory or text file. ## BC-breaking (Optional) Not that I am aware of. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. --------- Co-authored-by: 谢昕辰 <xiexinch@outlook.com> Co-authored-by: CSH <40987381+csatsurnh@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 5ce7667 - Browse repository at this point
Copy the full SHA 5ce7667View commit details
Commits on May 15, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 5459d34 - Browse repository at this point
Copy the full SHA 5459d34View commit details
Commits on May 18, 2023
-
Configuration menu - View commit details
-
Copy full SHA for cfd92e4 - Browse repository at this point
Copy the full SHA cfd92e4View commit details
Commits on May 22, 2023
-
[Project] Added a supported for Visual Attention Network (VAN) (open-…
…mmlab#2987) ## Motivation The original version of Visual Attention Network (VAN) can be found from https://github.com/Visual-Attention-Network/VAN-Segmentation 添加Visual Attention Network (VAN)的支持。 ## Modification added a floder mmsegmentation/projects/van/ added 13 configs totally and aligned performance basically. 只增加了一个文件夹,共增加13个配置文件,基本对齐性能(没有全部跑)。 ## Use cases (Optional) Before running, you may need to download the pretrain model from https://cloud.tsinghua.edu.cn/d/0100f0cea37d41ba8d08/ and then move them to the folder mmsegmentation/pretrained/, i.e. "mmsegmentation/pretrained/van_b2.pth". After that, run the following command: cd mmsegmentation bash tools/dist_train.sh projects/van/configs/van/van-b2_pre1k_upernet_4xb2-160k_ade20k-512x512.py 4 --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 2fb6d04 - Browse repository at this point
Copy the full SHA 2fb6d04View commit details
Commits on May 23, 2023
-
Configuration menu - View commit details
-
Copy full SHA for f886d86 - Browse repository at this point
Copy the full SHA f886d86View commit details
Commits on May 30, 2023
-
[Doc] Repair invalid link of potsdam and vaihingen (open-mmlab#3042)
Repair invalid link of potsdam and vaihingen in docs
Configuration menu - View commit details
-
Copy full SHA for 447537b - Browse repository at this point
Copy the full SHA 447537bView commit details -
Fixes an issue in isaid.py (open-mmlab#3010)
## Motivation When processing data in the isaid experiment, generated images only have binary pixel values of 0 or 1 instead of the corresponding class values. This causes significant interference and prevents subsequent experiments from proceeding. After investigation, it was found that the issue was caused by using the wrong image format for saving the images. Saving the images as PNG resulted in binary pixel values, while saving the images as BMP resolved the issue and correctly saved the class values. ## Modification `img_patch.save(save_path_image, format='BMP')` This code will save the image data as BMP format. ## BC-breaking Confirm that the modification does not introduce new issues and test that the modified code successfully resolves the original problem. --------- Co-authored-by: 谢昕辰 <xiexinch@outlook.com> Co-authored-by: CSH <40987381+csatsurnh@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 05014b0 - Browse repository at this point
Copy the full SHA 05014b0View commit details
Commits on Jun 5, 2023
-
[Dataset] Support GID dataset on project (open-mmlab#3038)
## Motivation Support GID dataset on project --------- Co-authored-by: 谢昕辰 <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 71878cc - Browse repository at this point
Copy the full SHA 71878ccView commit details
Commits on Jun 6, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 42ad908 - Browse repository at this point
Copy the full SHA 42ad908View commit details -
Configuration menu - View commit details
-
Copy full SHA for 4b69c40 - Browse repository at this point
Copy the full SHA 4b69c40View commit details -
[Fix] Fix bug cannot use both '--tta' and '--out' while testing. (ope…
…n-mmlab#3067) ## Motivation Fix bug cannot use both '--tta' and '--out' while testing. For details, please refer to open-mmlab#3064 . ## Modification Add 'img_path' in TTA predictions. --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 4f3b01e - Browse repository at this point
Copy the full SHA 4f3b01eView commit details
Commits on Jun 7, 2023
-
[Docs] Add docs contents at README.md (open-mmlab#3083)
Add docs contents at README.md to easily find documents. Issue: open-mmlab#2664 ![image](https://github.com/open-mmlab/mmsegmentation/assets/50650583/0763e108-f095-44d1-8e2a-ba2e83f02625) ![image](https://github.com/open-mmlab/mmsegmentation/assets/50650583/088e0945-55b4-4d3f-97bb-02e39dfafb5e) ![image](https://github.com/open-mmlab/mmsegmentation/assets/50650583/109e0320-d6a9-405f-b169-87acb5a0f94d) ![image](https://github.com/open-mmlab/mmsegmentation/assets/50650583/d1d72bde-40c0-451a-be6e-010f1f2d193b)
Configuration menu - View commit details
-
Copy full SHA for 9ca917f - Browse repository at this point
Copy the full SHA 9ca917fView commit details
Commits on Jun 16, 2023
-
[Docker] update Dockerfile libgl1-mesa-dev (open-mmlab#3095)
add libgl1-mesa-dev --------- Co-authored-by: 谢昕辰 <xiexinch@outlook.com> Co-authored-by: CSH <40987381+csatsurnh@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 2779df1 - Browse repository at this point
Copy the full SHA 2779df1View commit details -
[Feature] enhance swin pretrained model loading (open-mmlab#3097)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Enhance pretrained SwinTransformer loading when setting non-standard backbone `depths`. ## Modification Enhance pretrained SwinTransformer loading when setting non-standard backbone `depths`. ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials. --------- Co-authored-by: SheffieldCao <1751899@tongji.edu.cn>
Configuration menu - View commit details
-
Copy full SHA for 63e730b - Browse repository at this point
Copy the full SHA 63e730bView commit details -
[Fix] Robust mapping from image path to seg map path (open-mmlab#3091)
## Motivation Suppose an image is named `jpg.jpg` and its corresponding segmap is named `jpg.png`. The original implementation will try to read segmap from `png.png` and causes FileNotfoundError ## Modification Only replace the suffix, instead of full string search and replacement. ## BC-breaking (Optional) Probably no. ## Use cases (Optional) ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials. --------- Co-authored-by: 谢昕辰 <xiexinch@outlook.com> Co-authored-by: CSH <40987381+csatsurnh@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 1b7babe - Browse repository at this point
Copy the full SHA 1b7babeView commit details
Commits on Jun 19, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 49d9da9 - Browse repository at this point
Copy the full SHA 49d9da9View commit details -
[Feature] huasdorff distance loss (open-mmlab#2820)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Add Huasdorff distance loss --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for ab015da - Browse repository at this point
Copy the full SHA ab015daView commit details -
Configuration menu - View commit details
-
Copy full SHA for db2c210 - Browse repository at this point
Copy the full SHA db2c210View commit details -
[Enhancement] Change assertion logic inference cfg.model.test_cfg (op…
…en-mmlab#3012) ## Motivation In encode_decoder.py , assertion logic is not working correctly if user modifes cfg.test_cfg and defines it in a dictionary format. See: open-mmlab#3011 ## Modification Slight change to assertion behaviour to change assertion depending on if received test_cfg object is a dict or not. ## BC-breaking (Optional) Unsure - I believe this will not break any downstream tasks as the previous logic is still included ## Use cases (Optional) n/a --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 2fc1726 - Browse repository at this point
Copy the full SHA 2fc1726View commit details -
Configuration menu - View commit details
-
Copy full SHA for dc66230 - Browse repository at this point
Copy the full SHA dc66230View commit details
Commits on Jun 20, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 9271f91 - Browse repository at this point
Copy the full SHA 9271f91View commit details -
Configuration menu - View commit details
-
Copy full SHA for bbc916f - Browse repository at this point
Copy the full SHA bbc916fView commit details -
Configuration menu - View commit details
-
Copy full SHA for 0c6bf32 - Browse repository at this point
Copy the full SHA 0c6bf32View commit details -
Configuration menu - View commit details
-
Copy full SHA for a1d0756 - Browse repository at this point
Copy the full SHA a1d0756View commit details -
Configuration menu - View commit details
-
Copy full SHA for 8a72a2a - Browse repository at this point
Copy the full SHA 8a72a2aView commit details
Commits on Jun 21, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 583a2cb - Browse repository at this point
Copy the full SHA 583a2cbView commit details -
Configuration menu - View commit details
-
Copy full SHA for c941130 - Browse repository at this point
Copy the full SHA c941130View commit details -
Configuration menu - View commit details
-
Copy full SHA for ffb05dc - Browse repository at this point
Copy the full SHA ffb05dcView commit details -
Configuration menu - View commit details
-
Copy full SHA for c68c7ef - Browse repository at this point
Copy the full SHA c68c7efView commit details
Commits on Jun 22, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 4ef34d0 - Browse repository at this point
Copy the full SHA 4ef34d0View commit details
Commits on Jun 25, 2023
-
Configuration menu - View commit details
-
Copy full SHA for fa62b73 - Browse repository at this point
Copy the full SHA fa62b73View commit details -
Configuration menu - View commit details
-
Copy full SHA for 5057a5f - Browse repository at this point
Copy the full SHA 5057a5fView commit details -
Configuration menu - View commit details
-
Copy full SHA for 6123f1d - Browse repository at this point
Copy the full SHA 6123f1dView commit details -
Configuration menu - View commit details
-
Copy full SHA for 2972482 - Browse repository at this point
Copy the full SHA 2972482View commit details -
Configuration menu - View commit details
-
Copy full SHA for 086a4ad - Browse repository at this point
Copy the full SHA 086a4adView commit details -
Configuration menu - View commit details
-
Copy full SHA for 5141753 - Browse repository at this point
Copy the full SHA 5141753View commit details -
Configuration menu - View commit details
-
Copy full SHA for a86d9ed - Browse repository at this point
Copy the full SHA a86d9edView commit details -
Configuration menu - View commit details
-
Copy full SHA for a1445e0 - Browse repository at this point
Copy the full SHA a1445e0View commit details -
Configuration menu - View commit details
-
Copy full SHA for 6f6703e - Browse repository at this point
Copy the full SHA 6f6703eView commit details -
[Project] Medical semantic seg dataset: Chest x ray images with pneum…
…othorax masks (open-mmlab#2687)
Configuration menu - View commit details
-
Copy full SHA for 4fc70b9 - Browse repository at this point
Copy the full SHA 4fc70b9View commit details
Commits on Jun 26, 2023
-
[Fix] Fix dependency (open-mmlab#3136)
## Motivation Change the dependency `mmcls` to `mmpretrain` ## Modification - modify `mmcls` to `mmpretrain` - modify CI requirements ## BC-breaking (Optional) If users have installed mmcls but not install mmpretrain, it might raise some error.
Configuration menu - View commit details
-
Copy full SHA for a8d91dd - Browse repository at this point
Copy the full SHA a8d91ddView commit details -
Configuration menu - View commit details
-
Copy full SHA for 021ceac - Browse repository at this point
Copy the full SHA 021ceacView commit details
Commits on Jun 28, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 00c1568 - Browse repository at this point
Copy the full SHA 00c1568View commit details -
Configuration menu - View commit details
-
Copy full SHA for a13ca69 - Browse repository at this point
Copy the full SHA a13ca69View commit details -
[Feature] support mim download dataset (open-mmlab#3089)
## Motivation Please describe the motivation of this PR and the goal you want to achieve through this PR. ## Modification - add dataset-index.yml ## Dependencies - [ ] open-mmlab/mim#212
Configuration menu - View commit details
-
Copy full SHA for 50fae99 - Browse repository at this point
Copy the full SHA 50fae99View commit details
Commits on Jun 30, 2023
-
Configuration menu - View commit details
-
Copy full SHA for b949cc6 - Browse repository at this point
Copy the full SHA b949cc6View commit details
Commits on Jul 3, 2023
-
[Fix] Fix visualizor (open-mmlab#3154)
## Motivation **Current visualize result** ![rs-dev](https://github.com/open-mmlab/mmsegmentation/assets/15952744/147ea3f7-f632-457b-b257-031199320825) **Fixed the visualization result** ![rs-fix](https://github.com/open-mmlab/mmsegmentation/assets/15952744/98a86025-5a1e-4c2b-83e0-653dd659ba79) ## Modification remove mmengine `draw_binary_masks` api
Configuration menu - View commit details
-
Copy full SHA for 53b9922 - Browse repository at this point
Copy the full SHA 53b9922View commit details
Commits on Jul 4, 2023
-
Co-authored-by: CSH <40987381+csatsurnh@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 797f9fd - Browse repository at this point
Copy the full SHA 797f9fdView commit details -
Configuration menu - View commit details
-
Copy full SHA for e81b0f3 - Browse repository at this point
Copy the full SHA e81b0f3View commit details
Commits on Jul 12, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 9166f89 - Browse repository at this point
Copy the full SHA 9166f89View commit details
Commits on Jul 13, 2023
-
Configuration menu - View commit details
-
Copy full SHA for ac3a546 - Browse repository at this point
Copy the full SHA ac3a546View commit details -
[Fix] Fix SegTTAModel with no attribute '_gt_sem_seg' error (open-mml…
…ab#3152) ## Motivation When using the - tta command for multi-scale prediction, and the test set is not annotated, although format_only has been set true in test_evaluator, but SegTTAModel class still threw error 'AttributeError: 'SegDataSample' object has no attribute '_gt_sem_seg''. ## Modification The reason is SegTTAModel didn't determine if there were annotations in the dataset, so I added the code to make the judgment and let the program run normally on my computer.
Configuration menu - View commit details
-
Copy full SHA for cb376fe - Browse repository at this point
Copy the full SHA cb376feView commit details
Commits on Jul 14, 2023
-
[Feature] add bdd100K datasets (open-mmlab#3158)
## Motivation Integrate [BDD100K](https://paperswithcode.com/dataset/bdd100k) dataset. It shares the same classes as Cityscapes, and it's commonly used for evaluating segmentation/detection tasks in driving scenes, such as in [RobustNet](https://arxiv.org/abs/2103.15597), [WildNet](https://github.com/suhyeonlee/WildNet). Enhancement for Add BDD100K Dataset open-mmlab#2808 --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 2373bec - Browse repository at this point
Copy the full SHA 2373becView commit details -
Configuration menu - View commit details
-
Copy full SHA for 2a34ff3 - Browse repository at this point
Copy the full SHA 2a34ff3View commit details -
use correct link addresses for datasets in README_zh-CN (open-mmlab#3174
Configuration menu - View commit details
-
Copy full SHA for 7f5c416 - Browse repository at this point
Copy the full SHA 7f5c416View commit details -
[Fix] Albumentations default key mapping mismatch (open-mmlab#3195)
## Modification Fix Albumentations default key mapping mismatch as mentioned in [issue # 3179](open-mmlab#3179) by changing `self.keymap_to_albu = { 'img': 'image', 'gt_masks': 'masks'}` to `self.keymap_to_albu = { 'img': 'image', 'gt_seg_map': 'mask'}` ## Use cases (Optional) Example albu config ``` crop_size = (512, 512) albu_train_transforms = [ dict( type='PadIfNeeded', min_height=crop_size[0]*2, min_width=crop_size[1]*2, border_mode=0, always_apply=True), dict(type='Flip', always_apply=True), dict(type='Rotate', limit=(-180, 180), interpolation=4, always_apply=True), dict(type='RandomScale', scale_limit=0.1, interpolation=4, always_apply=True), dict( type='ElasticTransform', alpha=20, sigma=15, interpolation=4, border_mode=0, mask_value=(0, 0, 0), approximate=True, same_dxdy=True, p=0.8), dict(type='ColorJitter', brightness=0.2, contrast=0.1, saturation=0.2, hue=0.2, always_apply=True), dict(type='AdvancedBlur', p=0.5), dict(type='CenterCrop', height=crop_size[0], width=crop_size[1], always_apply=True) ] ``` Example training pipeline without specifying `keymap` ``` train_pipeline = [ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', reduce_zero_label=False), dict( type='Albu', transforms=albu_train_transforms, ), dict(type='Resize', scale=crop_size, keep_ratio=False, interpolation='lanczos'), dict(type='PackSegInputs') ] ``` Example viz_dataset before the issue fixing ![A thaliana Lucia 07 2022 col-0 30m-1h after Infection 22162057 object id 0](https://github.com/open-mmlab/mmsegmentation/assets/66273343/5431472a-83fd-485f-aeb7-c65f27f1993d) Example viz_dataset after the issue fixing ![A thaliana Lucia 07 2022 col-0 30m-1h after Infection 22162057 object id 0](https://github.com/open-mmlab/mmsegmentation/assets/66273343/3d6d4937-41f0-4a18-ae47-35bc43d78843) --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 1040563 - Browse repository at this point
Copy the full SHA 1040563View commit details
Commits on Jul 20, 2023
-
[Enhancement] Remove batch inference assertion (open-mmlab#3210)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation open-mmlab#3181 open-mmlab#2965 open-mmlab#2644 open-mmlab#1645 open-mmlab#1444 open-mmlab#1370 open-mmlab#125 ## Modification Remove the assertion at data_preprocessor ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for 906b6ec - Browse repository at this point
Copy the full SHA 906b6ecView commit details
Commits on Jul 24, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 331f79c - Browse repository at this point
Copy the full SHA 331f79cView commit details
Commits on Aug 1, 2023
-
[CodeCamp2023-154] Add semantic label to the segmentation visualizati…
…on results (open-mmlab#3229) Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation [Add semantic label to the segmentation visualization results 分割可视化结果中加上语义信息 open-mmlab#154](open-mmlab/OpenMMLabCamp#154) corresponding issue: [跑出来结果之后怎么在结果图片上获取各个语意部分的区域信息? open-mmlab#2578](open-mmlab#2578) ## Modification 1. mmseg/apis/inference.py, add withLabels in visualizer.add_datasample call, to indicate whether add semantic label 2. mmseg/visualization/local_visualizer.py, add semantic labels by opencv; modify the demo comment description 3. mmseg/utils/__init__.py, add bdd100k datasets to test local_visualizer.py **Current visualize result** <img width="637" alt="image" src="https://github.com/open-mmlab/mmsegmentation/assets/35064479/6ef6ce02-1d82-46f8-bde9-a1d69ff62df8"> **Add semantic label** <img width="637" alt="image" src="https://github.com/open-mmlab/mmsegmentation/assets/35064479/00716679-b43a-4794-8499-9bfecdb4b78b"> ## Test results **tests/test_visualization/test_local_visualizer.py** test results:(MMSegmentation/tests/data/pseudo_cityscapes_dataset/leftImg8bit/val/frankfurt/frankfurt_000000_000294_leftImg8bit.png) <img width="643" alt="image" src="https://github.com/open-mmlab/mmsegmentation/assets/35064479/6792b7d2-2512-4ea9-8500-1a7ed2d5e0dc"> **demo/inference_demo.ipynb** test results: <img width="966" alt="image" src="https://github.com/open-mmlab/mmsegmentation/assets/35064479/dfc0147e-fb1a-490a-b6ff-a8b209352d9b"> ----- ## Drawbacks config opencv thickness according to image size <img width="496" alt="image" src="https://github.com/open-mmlab/mmsegmentation/assets/35064479/0a54d72c-62b1-422c-89ae-69dc753fe0fc"> I have no idea of dealing with label overlapping for the time being
Configuration menu - View commit details
-
Copy full SHA for 33b87ec - Browse repository at this point
Copy the full SHA 33b87ecView commit details
Commits on Aug 9, 2023
-
[Fix] Fix module PascalContextDataset (open-mmlab#3235)
## Motivation - 'PascalContextDataset' object has no attribute 'file_client', it will cause an error. - The attribute ‘ann_file’ is not allowed to be empty, otherwise, an error will be reported. ## Modification - Replace file_client with fileio
Configuration menu - View commit details
-
Copy full SHA for d849ce7 - Browse repository at this point
Copy the full SHA d849ce7View commit details -
[Fix] Added ignore_index and one hot encoding for dice loss (open-mml…
…ab#3237) Added ignore_index param to forward(), also implemented one hot encoding to ensure the dims of target matches pred. Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Please describe the motivation of this PR and the goal you want to achieve through this PR. Attempted to solve the problems mentioned by open-mmlab#3172 ## Modification Please briefly describe what modification is made in this PR. Added ignore_index into forward function (although the dice loss itself does not actually take account for it for some reason). Added _expand_onehot_labels_dice, which takes the target with shape [N, H, W] into [N, num_classes, H, W]. ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials. This is my first time contributing to open-source code, so I might have made some stupid mistakes. Please don't hesitate to point it out.
Configuration menu - View commit details
-
Copy full SHA for f0a0e21 - Browse repository at this point
Copy the full SHA f0a0e21View commit details -
Configuration menu - View commit details
-
Copy full SHA for 0b67c31 - Browse repository at this point
Copy the full SHA 0b67c31View commit details -
[Project] Support CAT-Seg from CVPR2023 (open-mmlab#3098)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Support CAT-Seg open-vocabulary semantic segmentation (CVPR2023). ## Modification Support CAT-Seg open-vocabulary semantic segmentation (CVPR2023). - [x] Support CAT-Seg model training. - [x] CLIP model based `backbone` (R101 & Swin-B), aggregation layers based `neck`, and `decoder` head. - [x] Provide customized coco-stuff164k_384x384 training configs. - [x] Language model supports for `open vocabulary` (OV) tasks. - [x] Support CLIP-based pretrained language model (LM) inference. - [x] Add commonly used prompts templates. - [x] Add README tutorials. - [x] Add zero-shot testing scripts. **Working on the following tasks.** - [x] Add unit test. ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials. --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 611779d - Browse repository at this point
Copy the full SHA 611779dView commit details
Commits on Aug 10, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 5e0a45d - Browse repository at this point
Copy the full SHA 5e0a45dView commit details
Commits on Aug 15, 2023
-
[Doc]fix inference_segmentor to inference_model (open-mmlab#3261)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation there is a code mistake in `docs\zh_cn\get_started.md` and `docs\en\get_started.md`,it use the 0.x api,which is changed in 1.x ## Modification `docs\zh_cn\get_started.md` ,`docs\en\get_started.md` fix inference_segmentor --> inference_model ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for 5e673c6 - Browse repository at this point
Copy the full SHA 5e673c6View commit details
Commits on Aug 16, 2023
-
Configuration menu - View commit details
-
Copy full SHA for aa75916 - Browse repository at this point
Copy the full SHA aa75916View commit details -
[Project] Add pp_mobileseg onnx inference demo (open-mmlab#3268)
## Motivation Add a model deployment example. ## Modification Add an inference script and update the README. ## BC-breaking (Optional) None ## Use cases (Optional) In README.
Configuration menu - View commit details
-
Copy full SHA for a02f4c6 - Browse repository at this point
Copy the full SHA a02f4c6View commit details
Commits on Aug 17, 2023
-
[Feature] Support NYU depth estimation dataset (open-mmlab#3269)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Please describe the motivation of this PR and the goal you want to achieve through this PR. ## Modification Please briefly describe what modification is made in this PR. 1. add `NYUDataset`class 2. add script to process NYU dataset 3. add transforms for loading depth map 4. add docs & unittest ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 5. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 6. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 7. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for 8193195 - Browse repository at this point
Copy the full SHA 8193195View commit details
Commits on Aug 23, 2023
-
Configuration menu - View commit details
-
Copy full SHA for b66d7ae - Browse repository at this point
Copy the full SHA b66d7aeView commit details
Commits on Aug 25, 2023
-
[Doc] translate doc for docs/zh_cn/user_guides/5_deployment.md (open-…
…mmlab#3281) Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation translate doc for docs/zh_cn/user_guides/5_deployment.md ## Modification update `docs/en/user_guides/5_deployment.md` fix `docs/zh_cn/user_guides/5_deployment.md` ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials. --------- Co-authored-by: 谢昕辰 <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 561e44d - Browse repository at this point
Copy the full SHA 561e44dView commit details
Commits on Aug 28, 2023
-
[CodeCamp2023-526] Kullback-Leibler divergence Loss implementation (o…
…pen-mmlab#3242) Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation It's OpenMMLab Codecamp task. ## Modification Implementd Kullback-Leibler divergence loss and also added tests for it. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials. --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 3f11820 - Browse repository at this point
Copy the full SHA 3f11820View commit details
Commits on Aug 31, 2023
-
[Feature] Support depth metrics (open-mmlab#3297)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Please describe the motivation of this PR and the goal you want to achieve through this PR. Support metrics for the depth estimation task, including RMSE, ABSRel, and etc. ## Modification Please briefly describe what modification is made in this PR. ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) Using the following configuration to compute depth metrics on NYU ```python dataset_type = 'NYUDataset' data_root = 'data/nyu' test_pipeline = [ dict(type='LoadImageFromFile'), dict(dict(type='LoadDepthAnnotation', depth_rescale_factor=1e-3)), dict( type='PackSegInputs', meta_keys=('img_path', 'depth_map_path', 'ori_shape', 'img_shape', 'pad_shape', 'scale_factor', 'flip', 'flip_direction', 'category_id')) ] val_dataloader = dict( batch_size=1, num_workers=4, persistent_workers=True, sampler=dict(type='DefaultSampler', shuffle=False), dataset=dict( type=dataset_type, data_root=data_root, test_mode=True, data_prefix=dict( img_path='images/test', depth_map_path='annotations/test'), pipeline=test_pipeline)) test_dataloader = val_dataloader val_evaluator = dict(type='DepthMetric', max_depth_eval=10.0, crop_type='nyu') test_evaluator = val_evaluator ``` Example log: ![image](https://github.com/open-mmlab/mmsegmentation/assets/26127467/8101d65c-dee6-48de-916c-818659947b59) ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for a784422 - Browse repository at this point
Copy the full SHA a784422View commit details -
[Feature] remote sensing inference (open-mmlab#3131)
## Motivation Supports inference for ultra-large-scale remote sensing images. ## Modification Add RSImageInference.py in demo. ## Use cases Taking the inference of Vaihingen dataset images using PSPNet as an example, the following settings are required: **img**: Specify the path of the image. **model**: Provide the configuration file for the model. **checkpoint**: Specify the weight file for the model. **out**: Set the output path for the results. **batch_size**: Determine the batch size used during inference. **win_size**: Specify the width and height(512x512) of the sliding window. **stride**: Set the stride(400x400) for sliding the window. **thread(default: 1)**: Specify the number of threads to be used for inference. **Inference device (default: cuda:0)**: Specify the device for inference (e.g., cuda:0 for CPU). ```shell python demo/rs_image_inference.py demo/demo.png projects/pp_mobileseg/configs/pp_mobileseg/pp_mobileseg_mobilenetv3_2x16_80k_ade20k_512x512_tiny.py pp_mobileseg_mobilenetv3_2xb16_3rdparty-tiny_512x512-ade20k-a351ebf5.pth --batch-size 8 --device cpu --thread 2 ``` --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for f8f0151 - Browse repository at this point
Copy the full SHA f8f0151View commit details -
[Fix] Update confusion_matrix.py (open-mmlab#3291)
## Motivation ## Modification The confusion_matrix.py is not compatible with the current version of mmseg. --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 2bf6b6e - Browse repository at this point
Copy the full SHA 2bf6b6eView commit details
Commits on Sep 13, 2023
-
[Feature] Support VPD Depth Estimator (open-mmlab#3321)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Support depth estimation algorithm [VPD](https://github.com/wl-zhao/VPD) ## Modification 1. add VPD backbone 2. add VPD decoder head for depth estimation 3. add a new segmentor `DepthEstimator` based on `EncoderDecoder` for depth estimation 4. add an integrated metric that calculate common metrics in depth estimation 5. add SiLog loss for depth estimation 6. add config for VPD ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 7. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 8. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 9. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for eee780f - Browse repository at this point
Copy the full SHA eee780fView commit details -
Configuration menu - View commit details
-
Copy full SHA for 5a3869c - Browse repository at this point
Copy the full SHA 5a3869cView commit details
Commits on Sep 15, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 5bfe0d6 - Browse repository at this point
Copy the full SHA 5bfe0d6View commit details
Commits on Sep 18, 2023
-
Configuration menu - View commit details
-
Copy full SHA for bebeaf1 - Browse repository at this point
Copy the full SHA bebeaf1View commit details -
Configuration menu - View commit details
-
Copy full SHA for 2c106d0 - Browse repository at this point
Copy the full SHA 2c106d0View commit details -
[Feature] Support inference and visualization of VPD (open-mmlab#3331)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Support inference and visualization of VPD ## Modification 1. add a new VPD model that does not generate black border in predictions 2. update `SegLocalVisualizer` to support depth visualization 3. update `MMSegInferencer` to support save predictions of depth estimation in method `postprocess` ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) Run inference with VPD using the this command ```sh python demo/image_demo_with_inferencer.py demo/classroom__rgb_00283.jpg vpd_depth --out-dir vis_results ``` The following image will be saved under `vis_results/vis` ![classroom__rgb_00283](https://github.com/open-mmlab/mmsegmentation/assets/26127467/051e8c4b-8f92-495f-8c3e-f249aac888e3) ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 4. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 5. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 6. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for a50d071 - Browse repository at this point
Copy the full SHA a50d071View commit details
Commits on Sep 20, 2023
-
Use the pytorch-grad-cam tool to visualize Class Activation Maps (CAM) (
open-mmlab#3324) Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Use the pytorch-grad-cam tool to visualize Class Activation Maps (CAM). ## Modification Use the pytorch-grad-cam tool to visualize Class Activation Maps (CAM). requirement: pip install grad-cam run commad: python tools/analysis_tools/visualization_cam.py ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for 3db2f44 - Browse repository at this point
Copy the full SHA 3db2f44View commit details -
Configuration menu - View commit details
-
Copy full SHA for ecfc9aa - Browse repository at this point
Copy the full SHA ecfc9aaView commit details -
Configuration menu - View commit details
-
Copy full SHA for 27337e1 - Browse repository at this point
Copy the full SHA 27337e1View commit details -
[Feature] Support Side Adapter Network (open-mmlab#3232)
## Motivation Support SAN for Open-Vocabulary Semantic Segmentation Paper: [Side Adapter Network for Open-Vocabulary Semantic Segmentation](https://arxiv.org/abs/2302.12242) official Code: [SAN](https://github.com/MendelXu/SAN) ## Modification - Added the parameters of backbone vit for implementing the image encoder of CLIP. - Added text encoder code. - Added segmentor multimodel encoder-decoder code for open-vocabulary semantic segmentation. - Added SideAdapterNetwork decode head code. - Added config files for train and inference. - Added tools for converting pretrained models. - Added loss implementation for mask classification model, such as SAN, Maskformer and remove dependency on mmdetection. - Added test units for text encoder, multimodel encoder-decoder, san decode head and hungarian_assigner. ## Use cases ### Convert Models **pretrained SAN model** The official pretrained model can be downloaded from [san_clip_vit_b_16.pth](https://huggingface.co/Mendel192/san/blob/main/san_vit_b_16.pth) and [san_clip_vit_large_14.pth](https://huggingface.co/Mendel192/san/blob/main/san_vit_large_14.pth). Use tools/model_converters/san2mmseg.py to convert offcial model into mmseg style. `python tools/model_converters/san2mmseg.py <MODEL_PATH> <OUTPUT_PATH>` **pretrained CLIP model** Use the CLIP model provided by openai to train SAN. The CLIP model can be download from [ViT-B-16.pt](https://openaipublic.azureedge.net/clip/models/5806e77cd80f8b59890b7e101eabd078d9fb84e6937f9e85e4ecb61988df416f/ViT-B-16.pt) and [ViT-L-14-336px.pt](https://openaipublic.azureedge.net/clip/models/3035c92b350959924f9f00213499208652fc7ea050643e8b385c2dac08641f02/ViT-L-14-336px.pt). Use tools/model_converters/clip2mmseg.py to convert model into mmseg style. `python tools/model_converters/clip2mmseg.py <MODEL_PATH> <OUTPUT_PATH>` ### Inference test san_vit-base-16 model on coco-stuff164k dataset `python tools/test.py ./configs/san/san-vit-b16_coco-stuff164k-640x640.py <TRAINED_MODEL_PATH>` ### Train test san_vit-base-16 model on coco-stuff164k dataset `python tools/train.py ./configs/san/san-vit-b16_coco-stuff164k-640x640.py --cfg-options model.pretrained=<PRETRAINED_MODEL_PATH>` ## Comparision Results ### Train on COCO-Stuff164k | | | mIoU | mAcc | pAcc | | --------------- | ----- | ----- | ----- | ----- | | san-vit-base16 | official | 41.93 | 56.73 | 67.69 | | | mmseg | 41.93 | 56.84 | 67.84 | | san-vit-large14 | official | 45.57 | 59.52 | 69.76 | | | mmseg | 45.78 | 59.61 | 69.21 | ### Evaluate on Pascal Context | | | mIoU | mAcc | pAcc | | --------------- | ----- | ----- | ----- | ----- | | san-vit-base16 | official | 54.05 | 72.96 | 77.77 | | | mmseg | 54.04 | 73.74 | 77.71 | | san-vit-large14 | official | 57.53 | 77.56 | 78.89 | | | mmseg | 56.89 | 76.96 | 78.74 | ### Evaluate on Voc12Aug | | | mIoU | mAcc | pAcc | | --------------- | ----- | ----- | ----- | ----- | | san-vit-base16 | official | 93.86 | 96.61 | 97.11 | | | mmseg | 94.58 | 97.01 | 97.38 | | san-vit-large14 | official | 95.17 | 97.61 | 97.63 | | | mmseg | 95.58 | 97.75 | 97.79 | --------- Co-authored-by: CastleDream <35064479+CastleDream@users.noreply.github.com> Co-authored-by: yeedrag <46050186+yeedrag@users.noreply.github.com> Co-authored-by: Yang-ChangHui <71805205+Yang-Changhui@users.noreply.github.com> Co-authored-by: Xu CAO <49406546+SheffieldCao@users.noreply.github.com> Co-authored-by: xiexinch <xiexinch@outlook.com> Co-authored-by: 小飞猪 <106524776+ooooo-create@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 2ad007c - Browse repository at this point
Copy the full SHA 2ad007cView commit details
Commits on Sep 26, 2023
-
Configuration menu - View commit details
-
Copy full SHA for b6445a9 - Browse repository at this point
Copy the full SHA b6445a9View commit details -
Configuration menu - View commit details
-
Copy full SHA for 9f0c68c - Browse repository at this point
Copy the full SHA 9f0c68cView commit details -
Configuration menu - View commit details
-
Copy full SHA for d225ca7 - Browse repository at this point
Copy the full SHA d225ca7View commit details -
Configuration menu - View commit details
-
Copy full SHA for d88fba8 - Browse repository at this point
Copy the full SHA d88fba8View commit details
Commits on Oct 12, 2023
-
[CodeCamp2023-565] Fine tune ONNX Models (MMSegemetation) Inference f…
…or NVIDIA Jetson (open-mmlab#3372) Fine tune ONNX Models (MMSegemetation) Inference for NVIDIA Jetson
Configuration menu - View commit details
-
Copy full SHA for e090074 - Browse repository at this point
Copy the full SHA e090074View commit details -
Configuration menu - View commit details
-
Copy full SHA for 01e5ac1 - Browse repository at this point
Copy the full SHA 01e5ac1View commit details -
Configuration menu - View commit details
-
Copy full SHA for d30eaf9 - Browse repository at this point
Copy the full SHA d30eaf9View commit details -
Configuration menu - View commit details
-
Copy full SHA for efdddb1 - Browse repository at this point
Copy the full SHA efdddb1View commit details
Commits on Oct 16, 2023
-
[Fix] Add bpe_simple_vocab_16e6.txt.gz to release (open-mmlab#3386)
## Motivation open-mmlab#3383 ## Modification - Add bpe_simple_vocab_16e6.txt.gz to `MANIFEST.in`
Configuration menu - View commit details
-
Copy full SHA for 6450cd5 - Browse repository at this point
Copy the full SHA 6450cd5View commit details -
[Fix] Fix init api (open-mmlab#3388)
## Motivation open-mmlab#3384 ## Modification - mmseg/apis/inference.py
Configuration menu - View commit details
-
Copy full SHA for 3e49362 - Browse repository at this point
Copy the full SHA 3e49362View commit details
Commits on Oct 17, 2023
-
Configuration menu - View commit details
-
Copy full SHA for e3336d1 - Browse repository at this point
Copy the full SHA e3336d1View commit details
Commits on Dec 4, 2023
-
[Bugfix] Fix bug in cross entropy loss (open-mmlab#3457)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Fixes open-mmlab#3412 ## Modification We just need to replace tensor creation using torch.stack() instead of torch.tensor(). ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for 0df84b7 - Browse repository at this point
Copy the full SHA 0df84b7View commit details -
[Bugfix] Allow custom visualizer (open-mmlab#3455)
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation Current Visualization Hook can only get instances of `SegLocalVisualizer`. This makes impossible to use any other custom implementation. ## Modification This PR just allows to instantiate a different visualizer (following mmdetection implementation): https://github.com/open-mmlab/mmdetection/blob/main/mmdet/engine/hooks/visualization_hook.py#L58
Configuration menu - View commit details
-
Copy full SHA for 862a726 - Browse repository at this point
Copy the full SHA 862a726View commit details
Commits on Dec 7, 2023
-
[Bugfix] test resize with pad_shape (open-mmlab#3421)
## Motivation When using `test_cfg` for `data_preprocessor`, `predict_by_feat` resizes to the original size, not the padded size. ``` data_preprocessor = dict( type="SegDataPreProcessor", #type="SegDataPreProcessorWithPad", mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], bgr_to_rgb=True, pad_val=0, seg_pad_val=255, test_cfg=dict(size=(128, 128))) ``` Refar to: https://github.com/open-mmlab/mmsegmentation/blob/main/mmseg/models/decode_heads/san_head.py#L589-L592 ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials.
Configuration menu - View commit details
-
Copy full SHA for e4bfa6f - Browse repository at this point
Copy the full SHA e4bfa6fView commit details
Commits on Dec 14, 2023
-
[Feature] add -with-labels arg to inferencer for visualization withou…
…t labels (open-mmlab#3466) Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. ## Motivation It is difficult to visualize without "labels" when using the inferencer. - While using the `MMSegInferencer`, the visualized prediction contains labels on the mask, but it is difficult to pass `withLabels=False` without rewriting the config (which is harder to do when you initialize the inferencer with a model name rather than the config). - I thought it would be easier to just pass `withLabels=False` to `inferencer.__call__()` since you can also pass `opacity` and other parameters anyway. ## Modification Please briefly describe what modification is made in this PR. - Added `with_labels` to `visualize_kwargs` inside `MMSegInferencer`. - Modified to `visualize()` function. ## BC-breaking (Optional) Does the modification introduce changes that break the backward-compatibility of the downstream repos? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR. ## Use cases (Optional) If this PR introduces a new feature, it is better to list some use cases here, and update the documentation. ## Checklist 1. Pre-commit or other linting tools are used to fix the potential lint issues. 2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness. 3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D. 4. The documentation has been modified accordingly, like docstring or example tutorials. --------- Co-authored-by: xiexinch <xiexinch@outlook.com>
Configuration menu - View commit details
-
Copy full SHA for 3a2c835 - Browse repository at this point
Copy the full SHA 3a2c835View commit details -
Configuration menu - View commit details
-
Copy full SHA for 9d17509 - Browse repository at this point
Copy the full SHA 9d17509View commit details
Commits on Apr 4, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 04c9c64 - Browse repository at this point
Copy the full SHA 04c9c64View commit details
Commits on Apr 5, 2024
-
Configuration menu - View commit details
-
Copy full SHA for dff2ba6 - Browse repository at this point
Copy the full SHA dff2ba6View commit details -
Configuration menu - View commit details
-
Copy full SHA for f411286 - Browse repository at this point
Copy the full SHA f411286View commit details