Beyond FVD: An Enhanced Evaluation Metrics for Video Generation Distribution Quality
Ge Ya Luo, Gian Mario Favero, ZhiHao Luo, Alexia Jolicoeur-Martineau, Christopher Pal
Abstract
The Fréchet Video Distance (FVD) is a widely adopted metric for evaluating video generation distribution quality. However, its effectiveness relies on critical assumptions. Our analysis reveals three significant limitations: (1) the non-Gaussianity of the Inflated 3D Convnet (I3D) feature space; (2) the insensitivity of I3D features to temporal distortions; (3) the impractical sample sizes required for reliable estimation. These findings undermine FVD's reliability and show that FVD falls short as a standalone metric for video generation evaluation. After extensive analysis of a wide range of metrics and backbone architectures, we propose JEDi, the JEPA Embedding Distance, based on features derived from a Joint Embedding Predictive Architecture, measured using Maximum Mean Discrepancy with polynomial kernel. Our experiments on multiple open-source datasets show clear evidence that it is a superior alternative to the widely used FVD metric, requiring only 16% of the samples to reach its steady value, while increasing alignment with human evaluation by 34%, on average. Project page: https://oooolga.github.io/JEDi.github.io/.
BibTeX
@inproceedings{
luo2025beyond,
title={Beyond {FVD}: An Enhanced Evaluation Metrics for Video Generation Distribution Quality},
author={Ge Ya Luo and Gian Mario Favero and ZhiHao Luo and Alexia Jolicoeur-Martineau and Christopher Pal},
booktitle={The Thirteenth International Conference on Learning Representations},
year={2025},
url={https://openreview.net/forum?id=cC3LxGZasH}
}