action-segmentation

MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation (TPAMI 2020)

[CVPR 2020] Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation (PyTorch)

Official Repo for CVPR 2024 Paper "FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Fully-Supervised Action Segmentation"

Unsupervised learning of action classes with continuous temporal embedding (CVPR'19)

The official implementation of Error Detection in Egocentric Procedural Task Videos

DLC2Action is an action segmentation package that makes running and tracking of machine learning experiments easy.

Official Implementation for "Fast Weakly Supervised Action Segmentation Using Mutual Consistency" - TPAMI 2021

Simple vispy-based frame-by-frame behavior annotation GUI. Also can display DeepLabCut poses within videos during annotation.

U-shape Deep Networks are Unified Backbones for Human Action Understanding from Wi-Fi Signals

Official Repo for ICCV 2021 Paper "Weakly-Supervised Action Segmentation and Alignment via Transcript-Aware Union-of-Subspaces Learning".

Action Segmentation of Figure Skating Competition Videos: A Skeleton-Based Approach

Behavior Annotation Scores (BANOS) is a set of metrics for evaluating algorithmic behavior annotations against a ground truth (typically human annotations). It is designed for ethology and computer vision workflows where researchers annotate behaviors in video data.

Supervised Sliding Window Smoothing Loss Function Based on MS-TCN for Video Segmentation

MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation (TPAMI 2020) - adapted for surgical tool usage and added attention

A Medical Robotics project on Surgical Workflow Analysis, comparing deep learning architectures (GRU, MS-TCN, Transformers, SF-TMN) for surgical phase and instrument recognition.

Study Kodokan Judo throws from video via body-pose analysis

Research code from my NUS internship on the Tasty Videos dataset: ResNet-50/I3D features, S2VT captioning, and weakly supervised step segmentation (IEEE TPAMI 2022)

Deep learning approach to action segmentation in manufacturing assembly process.

A fast, unsupervised video segmentation tool using PyTorch. It detects action boundaries via feature similarity and refines segments with clustering, requiring no training or labels.