Rethinking FID: Towards a Better Evaluation Metric for Image Generation

By Sadeep Jayasumana, Srikumar Ramalingam, Andreas Veit, Daniel Glasner, Ayan Chakrabarti, Sanjiv Kumar. Computer Vision and Pattern Recognition, 2023

396 citations recorded by the source corpus.

As with many machine learning problems, the progress of image generation methods hinges on good evaluation metrics. One of the most popular is the Frechet Inception Distance (FID). FID estimates the distance between a distribution of Inception-v3 features of real images, and those of images generated by the algorithm. We highlight important drawbacks of FID: Inception's poor representation of the rich and varied content generated by modern text-to-image models, incorrect normality assumptions, and poor sample complexity. We call for a reevaluation of FID's use as the primary quality metric for generated images. We empirically demonstrate that FID contradicts human raters, it does not reflect gradual improvement of iterative text-to-image models, it does not capture distortion levels, and that it produces inconsistent results when varying the sample size. We also propose an alternative new metric, CMMD, based on richer CLIP embeddings and the maximum mean discrepancy distance with the Gaussian RBF kernel. It is an unbiased estimator that does not make any assumptions on the probability distribution of the embeddings and is sample efficient. Through extensive experiments and analysis, we demonstrate that FID-based evaluations of text-to-image models may be unreliable, and that CMMD of-fers a more robust and reliable assessment of image quality. A reference implementation of CMMD is available at: https://github.com/google-research/google-research/tree/master/cmmd.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/ee9017b716c49006f423595624593b87df0a491b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

  1. Lixin Xu
  2. Haohan Weng
  3. Si-Ya Liu
  4. Ze-Bao He
  5. Xinzhou Wang
  6. Fan Yang
  7. Zhuo Chen
  8. Yang-Dan Tao

Citations in the corpus