{"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/putting-the-object-back-into-video-object","title":"Putting the Object Back into Video Object Segmentation","arxiv_id":"2310.12982","date":"2023-10-19","proceeding":"CVPR 2024 1","authors":["Ho Kei Cheng","Seoung Wug Oh","Brian Price","Joon-Young Lee","Alexander Schwing"],"abstract":"We present Cutie, a video object segmentation (VOS) network with object-level memory reading, which puts the object representation from memory back into the video object segmentation result. Recent works on VOS employ bottom-up pixel-level memory reading which struggles due to matching noise, especially in the presence of distractors, resulting in lower performance in more challenging data. In contrast, Cutie performs top-down object-level memory reading by adapting a small set of object queries. Via those, it interacts with the bottom-up pixel features iteratively with a query-based object transformer (qt, hence Cutie). The object queries act as a high-level summary of the target object, while high-resolution feature maps are retained for accurate segmentation. Together with foreground-background masked attention, Cutie cleanly separates the semantics of the foreground object from the background. On the challenging MOSE dataset, Cutie improves by 8.7 J&F over XMem with a similar running time and improves by 4.2 J&F over DeAOT while being three times faster. Code is available at: https://hkchengrex.github.io/Cutie","url_abs":"https://arxiv.org/abs/2310.12982v2","url_pdf":"https://arxiv.org/pdf/2310.12982v2.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":"putting-the-object-back-into-video-object","repo_url":"https://github.com/hkchengrex/Cutie","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"semi-supervised-video-object-segmentation","task_name":"Semi-Supervised Video Object Segmentation"},{"task_slug":"video-object-segmentation","task_name":"Video Object Segmentation"},{"task_slug":"video-semantic-segmentation","task_name":"Video Semantic Segmentation"},{"task_slug":"visual-object-tracking","task_name":"Visual Object Tracking"}],"methods":[{"method_slug":"set","method_name":"SET"},{"method_slug":"vos","method_name":"VOS"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-23","task":"Semi-Supervised Video Object Segmentation","dataset":"BURST-test","model":"Cutie (base, MEGA, 600 pixels)","rank_in_archive_order":1,"of":2,"metrics":{"HOTA (all)":"66.0","HOTA (common)":"66.5","HOTA (uncommon)":"65.9"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-23","task":"Semi-Supervised Video Object Segmentation","dataset":"BURST-test","model":"Cutie (base, with mose, 600 pixels)","rank_in_archive_order":2,"of":2,"metrics":{"HOTA (all)":"62.6","HOTA (common)":"63.8","HOTA (uncommon)":"62.3"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-22","task":"Semi-Supervised Video Object Segmentation","dataset":"BURST-val","model":"Cutie (base, MEGA, 600 pixels)","rank_in_archive_order":1,"of":2,"metrics":{"HOTA (all)":"61.2","HOTA (common)":"65.0","HOTA (uncommon)":"60.3"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-22","task":"Semi-Supervised Video Object Segmentation","dataset":"BURST-val","model":"Cutie (base, with mose, 600 pixels)","rank_in_archive_order":2,"of":2,"metrics":{"HOTA (all)":"58.4","HOTA (common)":"61.8","HOTA (uncommon)":"57.5"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-1","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (test-dev)","model":"Cutie+ (base, MEGA)","rank_in_archive_order":1,"of":59,"metrics":{"F-measure (Mean)":"91.4","FPS":"17.9","J&F":"88.1","Jaccard (Mean)":"84.7"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-1","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (test-dev)","model":"Cutie (base, MEGA)","rank_in_archive_order":2,"of":59,"metrics":{"F-measure (Mean)":"89.9","FPS":"36.4","J&F":"86.1","Jaccard (Mean)":"82.4"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-1","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (test-dev)","model":"Cutie+ (base)","rank_in_archive_order":3,"of":59,"metrics":{"F-measure (Mean)":"89.2","FPS":"17.9","J&F":"85.9","Jaccard (Mean)":"82.6"},"uses_additional_data":false},{"leaderboard":"/sota/visual-object-tracking-on-davis-2017","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (val)","model":"Cutie+ (base)","rank_in_archive_order":2,"of":81,"metrics":{"F-measure (Mean)":"93.4","J&F":"90.5","Jaccard (Mean)":"87.5","Params(M)":"17.9"},"uses_additional_data":true},{"leaderboard":"/sota/visual-object-tracking-on-davis-2017","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (val)","model":"Cutie+ (base, MEGA)","rank_in_archive_order":8,"of":81,"metrics":{"F-measure (Mean)":"90.8","J&F":"88.1","Jaccard (Mean)":"85.5","Speed (FPS)":"17.9"},"uses_additional_data":true},{"leaderboard":"/sota/visual-object-tracking-on-davis-2017","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (val)","model":"Cutie (base)","rank_in_archive_order":10,"of":81,"metrics":{"F-measure (Mean)":"91.1","J&F":"87.9","Jaccard (Mean)":"84.6","Params(M)":"36.4"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"Cutie+ (base, MEGA)","rank_in_archive_order":2,"of":17,"metrics":{"F":"75.8","FPS":"17.9","J":"67.6","J&F":"71.7"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"Cutie+ (small, MEGA)","rank_in_archive_order":3,"of":17,"metrics":{"F":"74.5","FPS":"20.6","J":"66.0","J&F":"70.3"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"Cutie (base, MEGA)","rank_in_archive_order":4,"of":17,"metrics":{"F":"74.1","FPS":"36.4","J":"65.8","J&F":"69.9"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"Cutie (small, MEGA)","rank_in_archive_order":5,"of":17,"metrics":{"F":"72.9","FPS":"45.5","J":"64.3","J&F":"68.6"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"Cutie (base, with mose)","rank_in_archive_order":6,"of":17,"metrics":{"F":"72.3","FPS":"36.4","J":"64.2","J&F":"68.3"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"Cutie (small, with mose)","rank_in_archive_order":7,"of":17,"metrics":{"F":"71.7","FPS":"45.5","J":"63.1","J&F":"67.4"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"Cutie (base)","rank_in_archive_order":9,"of":17,"metrics":{"F":"67.9","FPS":"36.4","J":"60.0","J&F":"64.0"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-21","task":"Semi-Supervised Video Object Segmentation","dataset":"MOSE","model":"Cutie (small)","rank_in_archive_order":10,"of":17,"metrics":{"F":"66.2","FPS":"45.5","J":"58.2","J&F":"62.2"},"uses_additional_data":false},{"leaderboard":"/sota/video-object-segmentation-on-youtube-vos","task":"Semi-Supervised Video Object Segmentation","dataset":"YouTube-VOS 2018","model":"Cutie+ (base, MEGA)","rank_in_archive_order":1,"of":53,"metrics":{"F-Measure (Seen)":"91.0","F-Measure (Unseen)":"90.1","Jaccard (Seen)":"86.6","Jaccard (Unseen)":"82.2","Overall":"87.5","Speed (FPS)":"17.9"},"uses_additional_data":true},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-18","task":"Semi-Supervised Video Object Segmentation","dataset":"YouTube-VOS 2019","model":"Cutie+ (base, MEGA)","rank_in_archive_order":1,"of":22,"metrics":{"F-Measure (Seen)":"90.6","F-Measure (Unseen)":"90.5","J&F":"17.9","Jaccard (Seen)":"86.3","Jaccard (Unseen)":"82.7","Overall":"87.5"},"uses_additional_data":true},{"leaderboard":"/sota/video-object-segmentation-on-m-3-vos","task":"Video Object Segmentation","dataset":"M$^3$-VOS","model":"Cutie-base","rank_in_archive_order":2,"of":4,"metrics":{"Average IOU":"74.6"},"uses_additional_data":false},{"leaderboard":"/sota/video-object-segmentation-on-mose","task":"Video Object Segmentation","dataset":"MOSE","model":"Cutie","rank_in_archive_order":1,"of":1,"metrics":{"J&F":"68.3"},"uses_additional_data":true},{"leaderboard":"/sota/visual-object-tracking-on-didi","task":"Visual Object Tracking","dataset":"DiDi","model":"Cutie","rank_in_archive_order":6,"of":11,"metrics":{"Tracking quality":"0.575"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2310.12982","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.12982"}},"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/hkchengrex/Cutie","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":3,"unverified":2},"by_repo_kind":{"official":{"samples":5,"ran":3,"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":"066adf2df6def74d","entry":"ce_loss","repo":"hkchengrex/Cutie","repo_kind":"official","path":"cutie/model/losses.py","file_url":"https://github.com/hkchengrex/Cutie/blob/HEAD/cutie/model/losses.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"066adf2df6def74d"}},{"code_sha256_prefix":"0b4dc556327d69e6","entry":"dice_loss","repo":"hkchengrex/Cutie","repo_kind":"official","path":"cutie/model/losses.py","file_url":"https://github.com/hkchengrex/Cutie/blob/HEAD/cutie/model/losses.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0b4dc556327d69e6"}},{"code_sha256_prefix":"2048782602814ecb","entry":"downsample_groups","repo":"hkchengrex/Cutie","repo_kind":"official","path":"cutie/model/group_modules.py","file_url":"https://github.com/hkchengrex/Cutie/blob/HEAD/cutie/model/group_modules.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2048782602814ecb"}},{"code_sha256_prefix":"abf061dab6b3f9b9","entry":"interpolate_groups","repo":"hkchengrex/Cutie","repo_kind":"official","path":"cutie/model/group_modules.py","file_url":"https://github.com/hkchengrex/Cutie/blob/HEAD/cutie/model/group_modules.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"abf061dab6b3f9b9"}},{"code_sha256_prefix":"89405f8a57f3c3a5","entry":"upsample_groups","repo":"hkchengrex/Cutie","repo_kind":"official","path":"cutie/model/group_modules.py","file_url":"https://github.com/hkchengrex/Cutie/blob/HEAD/cutie/model/group_modules.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"89405f8a57f3c3a5"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}