The method projects the optimal-transport displacement underlying Fréchet distance onto a chosen feature direction and takes its expected squared magnitude.
Directional decomposition reveals what drives Fréchet scores across generative models
The authors break Fréchet distance into feature-space directions, exposing which visual, temporal and structural properties dominate the score.
Top University
Yunghee Lee · Jaeyeon Kim
Agency for Defense Development, South Korea · Harvard University
Research Digest··2 min read
Lee and Kim introduce directional Fréchet distance, which measures how much each feature-space direction contributes to the scalar distance between generated and reference distributions.
Why this paper
From Harvard University and Agency for Defense Development, South Korea
In one line
Directional Fréchet distance reveals that a few interpretable feature directions often dominate FID, FVD, and Protein FID discrepancies hidden by scalar scores.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§