songweige/content-debiased-fvd

[CVPR 2024] On the Content Bias in Fréchet Video Distance

36
/ 100
Emerging

This helps researchers and practitioners evaluate the realism and quality of videos generated by AI models. It takes a collection of generated videos and compares them to a set of real videos, providing a Fréchet Video Distance (FVD) score that more accurately reflects motion quality. This tool is for anyone developing or assessing video generation AI models, such as those in computer vision research or content creation.

144 stars. No commits in the last 6 months.

Use this if you need a more reliable and less 'content-biased' method to quantify how realistic your AI-generated videos are, particularly regarding their motion.

Not ideal if you are looking for a qualitative assessment or subjective user feedback on video quality, as this provides a quantitative metric.

video-generation AI-model-evaluation computer-vision deep-learning-research generative-AI
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

144

Forks

9

Language

Python

License

MIT

Last pushed

Sep 28, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/songweige/content-debiased-fvd"

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