Lightning-AI / Lightning-AI/torchmetrics

Frechet Inception Distance with Dinov2 features

Open
#3,110 8 comments 0 reactions 1 assignee View on GitHub

@gisilvs is already working on this.

Since Jul 4, 2025.

enhancement
Dominant language
Python
Stars
2.5k
Forks
526
Avg merge
6d 11h
Merged PRs (30d)
5

Description

## 🚀 Feature

Implement FID computation with DinoV2 as proposed in "Exposing flaws of generative model evaluation metrics
and their unfair treatment of diffusion models"

### Motivation

Computing FID with features extraded from DinoV2 has become common practice for generative models evaluation.

### Pitch

Either add a new class `FrechetInceptionDistanceDinoV2` based on torch-fidelity's `FeatureExtractorDinoV2`.

### Alternatives

Add an option in the existing `FrechetInceptionDistance `class to choose between inveption v3 and dino v2 feature extractor, also using the `FeatureExtractorDinoV2` from torch-fidelity.

### Additional context

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.