{"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/cut-and-learn-for-unsupervised-object","title":"Cut and Learn for Unsupervised Object Detection and Instance Segmentation","arxiv_id":"2301.11320","date":"2023-01-26","proceeding":"CVPR 2023 1","authors":["Xudong Wang","Rohit Girdhar","Stella X. Yu","Ishan Misra"],"abstract":"We propose Cut-and-LEaRn (CutLER), a simple approach for training unsupervised object detection and segmentation models. We leverage the property of self-supervised models to 'discover' objects without supervision and amplify it to train a state-of-the-art localization model without any human labels. CutLER first uses our proposed MaskCut approach to generate coarse masks for multiple objects in an image and then learns a detector on these masks using our robust loss function. We further improve the performance by self-training the model on its predictions. Compared to prior work, CutLER is simpler, compatible with different detection architectures, and detects multiple objects. CutLER is also a zero-shot unsupervised detector and improves detection performance AP50 by over 2.7 times on 11 benchmarks across domains like video frames, paintings, sketches, etc. With finetuning, CutLER serves as a low-shot detector surpassing MoCo-v2 by 7.3% APbox and 6.6% APmask on COCO when training with 5% labels.","url_abs":"https://arxiv.org/abs/2301.11320v1","url_pdf":"https://arxiv.org/pdf/2301.11320v1.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":"cut-and-learn-for-unsupervised-object","repo_url":"https://github.com/facebookresearch/cutler","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"cut-and-learn-for-unsupervised-object","repo_url":"https://github.com/u2seg/u2seg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"unsupervised-instance-segmentation","task_name":"Unsupervised Instance Segmentation"},{"task_slug":"unsupervised-object-detection","task_name":"Unsupervised Object Detection"},{"task_slug":"unsupervised-panoptic-segmentation","task_name":"Unsupervised Panoptic Segmentation"},{"task_slug":"unsupervised-zero-shot-instance-segmentation","task_name":"Unsupervised Zero-Shot Instance Segmentation"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/unsupervised-instance-segmentation-on-coco","task":"Unsupervised Instance Segmentation","dataset":"COCO val2017","model":"CutLER (Cascade+DINO)","rank_in_archive_order":2,"of":5,"metrics":{"AP":"9.2","AP50":"18.9","AP75":"9.7"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-instance-segmentation-on-uvo","task":"Unsupervised Instance Segmentation","dataset":"UVO","model":"CutLER (Cascade+DINO)","rank_in_archive_order":1,"of":1,"metrics":{"AP":"10.1","AP50":"22.8","AP75":"8"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-panoptic-segmentation-on-coco","task":"Unsupervised Panoptic Segmentation","dataset":"COCO val2017","model":"CutLER+STEGO","rank_in_archive_order":2,"of":2,"metrics":{"PQ":"12.4","RQ":"15.2","SQ":"36.1"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-zero-shot-instance-segmentation","task":"Unsupervised Zero-Shot Instance Segmentation","dataset":"COCO val2017","model":"CutLER","rank_in_archive_order":2,"of":2,"metrics":{"AP":"5.3","AP50":"8.6","AP75":"5.5","AR100":"9.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2301.11320","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}