line/Skeleton-Temporal-Action-Localization

Official implementation of AAAI 2023 Oral Paper "Frame-Level Label Refinement for Skeleton-Based Weakly-Supervised Action Recognition"

27
/ 100
Experimental

This project helps researchers and developers working with human motion data to automatically identify specific actions within long, unlabeled skeleton-based video recordings. You input raw human motion capture data (like AMASS and BABEL datasets) and it outputs precise, frame-level labels indicating when particular actions start and end. This is for machine learning researchers and computer vision engineers developing human activity recognition systems.

No commits in the last 6 months.

Use this if you need to precisely locate and label human actions in skeleton-based motion data for machine learning tasks, even with only high-level activity descriptions.

Not ideal if you are looking for a pre-trained, ready-to-use application for action recognition without extensive setup and dataset preparation.

human-activity-recognition motion-capture-analysis computer-vision-research action-localization machine-learning-research
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 6 / 25

How are scores calculated?

Stars

14

Forks

1

Language

Python

License

Apache-2.0

Last pushed

Oct 20, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/line/Skeleton-Temporal-Action-Localization"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.