{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/rethinking-fid-towards-a-better-evaluation","title":"Rethinking FID: Towards a Better Evaluation Metric for Image Generation","arxiv_id":"2401.09603","date":"2023-11-30","proceeding":"CVPR 2024 1","authors":["Sadeep Jayasumana","Srikumar Ramalingam","Andreas Veit","Daniel Glasner","Ayan Chakrabarti","Sanjiv Kumar"],"abstract":"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 offers a more robust and reliable assessment of image quality.","url_abs":"https://arxiv.org/abs/2401.09603v2","url_pdf":"https://arxiv.org/pdf/2401.09603v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"rethinking-fid-towards-a-better-evaluation","repo_url":"https://github.com/google-research/google-research","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"rethinking-fid-towards-a-better-evaluation","repo_url":"https://github.com/google-research/google-research/tree/master/cmmd","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"jax","reach":null},{"paper_slug":"rethinking-fid-towards-a-better-evaluation","repo_url":"https://github.com/mazurowski-lab/medical-image-similarity-metrics","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"rethinking-fid-towards-a-better-evaluation","repo_url":"https://github.com/sayakpaul/cmmd-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"auxiliary-classifier","method_name":"Auxiliary Classifier"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"clip","method_name":"CLIP"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"inception-v3","method_name":"Inception-v3"},{"method_slug":"inception-v3-module","method_name":"Inception-v3 Module"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"rbf","method_name":"RBF"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2401.09603","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2401.09603"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/google-research/google-research","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mazurowski-lab/medical-image-similarity-metrics","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/google-research/google-research/tree/master/cmmd","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sayakpaul/cmmd-pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":2,"unverified":1},"by_repo_kind":{"listed":{"samples":3,"ran":2,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"67af82f9a75b4a53","entry":"generate_images","repo":"sayakpaul/cmmd-pytorch","repo_kind":"listed","path":"generate_images.py","file_url":"https://github.com/sayakpaul/cmmd-pytorch/blob/HEAD/generate_images.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"67af82f9a75b4a53"}},{"code_sha256_prefix":"cbf43bceb31b386b","entry":"mmd","repo":"sayakpaul/cmmd-pytorch","repo_kind":"listed","path":"distance.py","file_url":"https://github.com/sayakpaul/cmmd-pytorch/blob/HEAD/distance.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cbf43bceb31b386b"}},{"code_sha256_prefix":"cb1471e7d1c79324","entry":"compute_embeddings_for_dir","repo":"sayakpaul/cmmd-pytorch","repo_kind":"listed","path":"io_util.py","file_url":"https://github.com/sayakpaul/cmmd-pytorch/blob/HEAD/io_util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cb1471e7d1c79324"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}