{"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/improving-visual-representation-learning","title":"Improving Visual Representation Learning through Perceptual Understanding","arxiv_id":"2212.14504","date":"2022-12-30","proceeding":"CVPR 2023 1","authors":["Samyakh Tukra","Frederick Hoffman","Ken Chatfield"],"abstract":"We present an extension to masked autoencoders (MAE) which improves on the representations learnt by the model by explicitly encouraging the learning of higher scene-level features. We do this by: (i) the introduction of a perceptual similarity term between generated and real images (ii) incorporating several techniques from the adversarial training literature including multi-scale training and adaptive discriminator augmentation. The combination of these results in not only better pixel reconstruction but also representations which appear to capture better higher-level details within images. More consequentially, we show how our method, Perceptual MAE, leads to better performance when used for downstream tasks outperforming previous methods. We achieve 78.1% top-1 accuracy linear probing on ImageNet-1K and up to 88.1% when fine-tuning, with similar results for other downstream tasks, all without use of additional pre-trained models or data.","url_abs":"https://arxiv.org/abs/2212.14504v2","url_pdf":"https://arxiv.org/pdf/2212.14504v2.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":"improving-visual-representation-learning","repo_url":"https://github.com/tractableai/perceptual-mae","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"self-supervised-image-classification","task_name":"Self-Supervised Image Classification"}],"methods":[{"method_slug":"mae","method_name":"MAE"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/self-supervised-image-classification-on","task":"Self-Supervised Image Classification","dataset":"ImageNet","model":"PercMAE (ViT-B, dVAE)","rank_in_archive_order":30,"of":144,"metrics":{"Number of Params":"80M","Top 1 Accuracy":"79.8%"},"uses_additional_data":true},{"leaderboard":"/sota/self-supervised-image-classification-on","task":"Self-Supervised Image Classification","dataset":"ImageNet","model":"PercMAE (ViT-B)","rank_in_archive_order":48,"of":144,"metrics":{"Number of Params":"80M","Top 1 Accuracy":"78.1%"},"uses_additional_data":false},{"leaderboard":"/sota/self-supervised-image-classification-on-1","task":"Self-Supervised Image Classification","dataset":"ImageNet (finetuned)","model":"PercMAE (ViT-L, dVAE)","rank_in_archive_order":2,"of":65,"metrics":{"Number of Params":"307M","Top 1 Accuracy":"88.6%"},"uses_additional_data":true},{"leaderboard":"/sota/self-supervised-image-classification-on-1","task":"Self-Supervised Image Classification","dataset":"ImageNet (finetuned)","model":"PercMAE (ViT-L)","rank_in_archive_order":5,"of":65,"metrics":{"Number of Params":"307M","Top 1 Accuracy":"88.1%"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}