create

Author: huangsiyu

.DS_Store

@@ -0,0 +1,92 @@
+BSD 3-Clause License
+
+Copyright (c) 2020, Siyu Huang
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this
+   list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright notice,
+   this list of conditions and the following disclaimer in the documentation
+   and/or other materials provided with the distribution.
+
+3. Neither the name of the copyright holder nor the names of its
+   contributors may be used to endorse or promote products derived from
+   this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+--------------------------- LICENSE FOR pytorch-CycleGAN-and-pix2pix --------------------------------
+For pytorch-CycleGAN-and-pix2pix software
+
+Copyright (c) 2017, Jun-Yan Zhu and Taesung Park
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+
+--------------------------- LICENSE FOR pix2pix --------------------------------
+BSD License
+
+For pix2pix software
+Copyright (c) 2016, Phillip Isola and Jun-Yan Zhu
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+
+----------------------------- LICENSE FOR DCGAN --------------------------------
+BSD License
+
+For dcgan.torch software
+
+Copyright (c) 2015, Facebook, Inc. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
+
+Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
+
+Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.
+
+Neither the name Facebook nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+

LICENSE

@@ -0,0 +1,120 @@
+# PoseStylizer
+PyTorch implementation of  "**Generating Person Images with Appearance-aware Pose Stylizer**" published in **IJCAI 2020**. 
+
+<p align="center"><img src='figs/framework.png' width="500"/></p>
+
+The code is written by [Siyu Huang](https://siyuhuang.github.io/). Contact [Siyu Huang](https://siyuhuang.github.io/) for questions on this project.
+
+<p align="center"><img src='figs/fig1.png' width="600"/></p>
+
+
+## Getting Started
+### Requirement
+* Python3
+* PyTorch 1.0.1
+
+### Installation
+- Clone this repo:
+```bash
+git clone https://github.com/siyuhuang/PoseStylizer.git
+cd PoseStylizer
+```
+
+### Data Preperation
+1. Download the Market-1501 dataset ```dataset/market_data.zip``` and the DeepFashion dataset ```dataset/fashion_data.zip``` from [Google Drive](https://drive.google.com/open?id=13EzWg6tW8a_DBabBbzCgTFuBqiuHahnu) / [Baidu Disk](https://pan.baidu.com/s/1PwO5yFhonDTtWdPLGgPrRw) (Password: jl0s).  The zip files include images ```/train``` ```/test```, keypoint annotations ```annotation.csv```, and pose transfer pairs ```pairs.csv```.
+```bash
+cd dataset
+unzip market_data.zip
+unzip fashion_data.zip
+cd ..
+```
+2. Generate the pose heatmaps. Note, the disk space of generated heatmaps are extremely large (~18GB for Market-1501 and ~160GB for DeepFashion).
+```bash
+python tool/generate_pose_map_market.py
+python tool/generate_pose_map_fashion.py
+```
+
+## Test with Pretrained Models
+
+Download our pretrained checkpoints from [Google Drive](https://drive.google.com/open?id=1LKFYZvwizRpDxslWoyUePC9XG0cpqLzT) / [Baidu Disk](https://pan.baidu.com/s/1PwO5yFhonDTtWdPLGgPrRw) (Password: jl0s). 
+* **Market-1501**
+
+```bash
+bash test_market.sh
+```
+
+* **DeepFashion**
+
+```bash
+bash test_fashion.sh
+```
+
+
+## Training
+* **Market-1501**
+```bash
+bash train_market.sh
+```
+
+* **DeepFashion**
+```bash
+bash train_fashion.sh
+```
+**Note:** We use 8 GPUs for training by default. If you have less GPUs, change ```--gpu_ids``` and ```--batchSize``` accordingly. The results are competitive to the results in our paper.
+
+## Evaluation
+
+### SSIM, IS, L1, mask-SSIM, mask-IS, mask-L1
+
+* *Tensorflow 1.14.1 (Python3)* is required. 
+
+* **Market-1501**
+```bash
+python tool/getMetrics_market.py
+```
+
+* **DeepFashion**
+```bash
+python tool/getMetrics_fashion.py
+```
+
+
+### PCKh
+* Download OpenPose pose estimator from [Google Drive](https://drive.google.com/open?id=1nqZ8xSXBXdL1F3WFHTJJwm4rXYCCVy0y) / [Baidu Disk](https://pan.baidu.com/s/1PwO5yFhonDTtWdPLGgPrRw) (Password: jl0s). Put ``
+pose_estimator.h5`` under the root folder ``PoseStylizer``.
+* *Tensorflow 1.14.1 (Python2)* is required. 
+1. **Crop generated images from ```/results``` folder.**
+```bash
+python tool/crop_market.py
+```
+&nbsp; &nbsp; or
+```bash
+python tool/crop_fashion.py
+```
+2. **Compute keypoints coordinates.** Change the paths ``input_folder``  and ``output_path`` in ``tool/compute_coordinates.py``. 
+```bash
+python2 tool/compute_coordinates.py
+```
+3. **Compute PCKh score.**
+```bash
+python tool/calPCKH_market.py
+```
+&nbsp; &nbsp; or
+```bash
+python tool/calPCKH_fashion.py
+```
+
+
+## Citation
+If you use this code for your research, please kindly cite our paper.
+```
+@inproceedings{huang2020generating,
+  title={Generating Person Images with Appearance-aware Pose Stylizer},
+  author={Huang, Siyu and Xiong, Haoyi and Cheng, Zhi-Qi and Wang, Qingzhong  and Zhou, Xingran and Wen, Bihan and Huan, Jun and Dou Dejing},
+  booktitle={IJCAI},
+  year={2020}
+}
+```
+
+## Acknowledgments
+The code is written based on [pytorch-CycleGAN-and-pix2pix](https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix) and [Pose-Transfer](https://github.com/tengteng95/Pose-Transfer).

README.md

@@ -0,0 +1,14 @@
+
+class BaseDataLoader():
+    def __init__(self):
+        pass
+    
+    def initialize(self, opt):
+        self.opt = opt
+        pass
+
+    def load_data():
+        return None
+
+        
+        

data/__init__.py

@@ -0,0 +1,43 @@
+import torch.utils.data as data
+from PIL import Image
+import torchvision.transforms as transforms
+
+class BaseDataset(data.Dataset):
+    def __init__(self):
+        super(BaseDataset, self).__init__()
+
+    def name(self):
+        return 'BaseDataset'
+
+    def initialize(self, opt):
+        pass
+
+def get_transform(opt):
+    transform_list = []
+    if opt.resize_or_crop == 'resize_and_crop':
+        osize = [opt.loadSize, opt.loadSize]
+        transform_list.append(transforms.Scale(osize, Image.BICUBIC))
+        transform_list.append(transforms.RandomCrop(opt.fineSize))
+    elif opt.resize_or_crop == 'crop':
+        transform_list.append(transforms.RandomCrop(opt.fineSize))
+    elif opt.resize_or_crop == 'scale_width':
+        transform_list.append(transforms.Lambda(
+            lambda img: __scale_width(img, opt.fineSize)))
+    elif opt.resize_or_crop == 'scale_width_and_crop':
+        transform_list.append(transforms.Lambda(
+            lambda img: __scale_width(img, opt.loadSize)))
+        transform_list.append(transforms.RandomCrop(opt.fineSize))
+
+    transform_list += [transforms.ToTensor(),
+                       transforms.Normalize((0.5, 0.5, 0.5),
+                                            (0.5, 0.5, 0.5))]
+    return transforms.Compose(transform_list)
+
+def __scale_width(img, target_width):
+    ow, oh = img.size
+    if (ow == target_width):
+        return img
+    w = target_width
+    h = int(target_width * oh / ow)
+    return img.resize((w, h), Image.BICUBIC)
+

data/base_data_loader.py

@@ -0,0 +1,45 @@
+import torch.utils.data
+from data.base_data_loader import BaseDataLoader
+
+
+def CreateDataset(opt):
+    dataset = None
+    
+    if opt.dataset_mode == 'keypoint':
+        from data.keypoint import KeyDataset
+        dataset = KeyDataset()
+    elif opt.dataset_mode == 'keypoint_segmentation':
+        from data.keypoint_segmentation import KeySegDataset
+        dataset = KeySegDataset()
+    else:
+        raise ValueError("Dataset [%s] not recognized." % opt.dataset_mode)
+
+    print("dataset [%s] was created" % (dataset.name()))
+    dataset.initialize(opt)
+    return dataset
+
+
+class CustomDatasetDataLoader(BaseDataLoader):
+    def name(self):
+        return 'CustomDatasetDataLoader'
+
+    def initialize(self, opt):
+        BaseDataLoader.initialize(self, opt)
+        self.dataset = CreateDataset(opt)
+        self.dataloader = torch.utils.data.DataLoader(
+            self.dataset,
+            batch_size=opt.batchSize,
+            shuffle=not opt.serial_batches,
+            num_workers=int(opt.nThreads))
+
+    def load_data(self):
+        return self
+
+    def __len__(self):
+        return min(len(self.dataset), self.opt.max_dataset_size)
+
+    def __iter__(self):
+        for i, data in enumerate(self.dataloader):
+            if i >= self.opt.max_dataset_size:
+                break
+            yield data

data/base_dataset.py

@@ -0,0 +1,7 @@
+
+def CreateDataLoader(opt):
+    from data.custom_dataset_data_loader import CustomDatasetDataLoader
+    data_loader = CustomDatasetDataLoader()
+    print(data_loader.name())
+    data_loader.initialize(opt)
+    return data_loader

data/custom_dataset_data_loader.py

@@ -0,0 +1,68 @@
+###############################################################################
+# Code from
+# https://github.com/pytorch/vision/blob/master/torchvision/datasets/folder.py
+# Modified the original code so that it also loads images from the current
+# directory as well as the subdirectories
+###############################################################################
+
+import torch.utils.data as data
+
+from PIL import Image
+import os
+import os.path
+
+IMG_EXTENSIONS = [
+    '.jpg', '.JPG', '.jpeg', '.JPEG',
+    '.png', '.PNG', '.ppm', '.PPM', '.bmp', '.BMP',
+]
+
+
+def is_image_file(filename):
+    return any(filename.endswith(extension) for extension in IMG_EXTENSIONS)
+
+
+def make_dataset(dir):
+    images = []
+    assert os.path.isdir(dir), '%s is not a valid directory' % dir
+
+    for root, _, fnames in sorted(os.walk(dir)):
+        for fname in fnames:
+            if is_image_file(fname):
+                path = os.path.join(root, fname)
+                images.append(path)
+
+    return images
+
+
+def default_loader(path):
+    return Image.open(path).convert('RGB')
+
+
+class ImageFolder(data.Dataset):
+
+    def __init__(self, root, transform=None, return_paths=False,
+                 loader=default_loader):
+        imgs = make_dataset(root)
+        if len(imgs) == 0:
+            raise(RuntimeError("Found 0 images in: " + root + "\n"
+                               "Supported image extensions are: " +
+                               ",".join(IMG_EXTENSIONS)))
+
+        self.root = root
+        self.imgs = imgs
+        self.transform = transform
+        self.return_paths = return_paths
+        self.loader = loader
+
+    def __getitem__(self, index):
+        path = self.imgs[index]
+        img = self.loader(path)
+        if self.transform is not None:
+            img = self.transform(img)
+        if self.return_paths:
+            return img, path
+        else:
+            return img
+
+    def __len__(self):
+        return len(self.imgs)