carlobretti/object-scene-compositions-for-actions

Code base for BMVC21 paper on zero-shot action recognition from object and scene compositions

25
/ 100
Experimental

This project helps computer vision researchers evaluate models that can recognize actions in videos without needing to be explicitly trained on those specific actions. It takes in video datasets, object and scene predictions for those videos, and action labels, then outputs action recognition predictions. A computer vision researcher or ML engineer working on video analysis would use this tool.

No commits in the last 6 months.

Use this if you are a computer vision researcher developing or evaluating zero-shot action recognition models and need a robust framework for testing performance.

Not ideal if you need a pre-trained, production-ready system for action recognition, as this is a research framework for model evaluation.

computer-vision action-recognition zero-shot-learning video-analysis machine-learning-research
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 4 / 25
Maturity 8 / 25
Community 13 / 25

How are scores calculated?

Stars

8

Forks

2

Language

Jupyter Notebook

License

Last pushed

Mar 25, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/carlobretti/object-scene-compositions-for-actions"

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