Video Query System
-
Updated
May 20, 2019 - Python
Video Query System
A PyTorch implementation of ACRNet based on ICME 2023 paper "Weakly-supervised Temporal Action Localization with Adaptive Clustering and Refining Network"
Localizing narrated human activities in lifestyle vlogs.
A curated list of papers in the domain of spatio-temporal action detection (or action localization.) NOTE: This repository does not contains papers for Temporal Action Localization.
[ICCV 2021] Official PyTorch implementation for "D2-Net: Weakly-Supervised Action Localization via Discriminative Embeddings and Denoised Activations"
ChangeIt dataset with more than 2600 hours of video with state-changing actions published at CVPR 2022
Temporal Action Localization Visualization Tool (TALVT) is a Javascript based simple visualization tool to visualize the outcomes of the code before using the evaluation script
Code for the paper "Multi-Task Learning of Object States and State-Modifying Actions from Web Videos" published in TPAMI
[BMVC 2021]: Official PyTorch implementation of : "Few Shot Temporal Action Localization using Query Adaptive Transformers"
Code for Look for the Change paper published at CVPR 2022
Tensorflow Reproduction of the EMNLP-2018 paper "Temporally Grounding Natural Sentence in Video"
Action-Localization, Atomic Visual Actions (AVA) Dataset
Adversarial Background-Aware Loss for Weakly-supervised Temporal Activity Localization (ECCV 2020)
Weakly-supervised Action Localization
Official Tensorflow Implementation of the AAAI-2020 paper "Temporally Grounding Language Queries in Videos by Contextual Boundary-aware Prediction"
Code for Spotting Temporally Precise, Fine-Grained Events in Video
[ICLR 2022] TAda! Temporally-Adaptive Convolutions for Video Understanding. This codebase provides solutions for video classification, video representation learning and temporal detection.
temporal action detection: benchmark results, features download etc.
Code release for ActionFormer (ECCV 2022)
Awesome video understanding toolkits based on PaddlePaddle. It supports video data annotation tools, lightweight RGB and skeleton based action recognition model, practical applications for video tagging and sport action detection.
Add a description, image, and links to the action-localization topic page so that developers can more easily learn about it.
To associate your repository with the action-localization topic, visit your repo's landing page and select "manage topics."