{"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/feelvos-fast-end-to-end-embedding-learning","title":"FEELVOS: Fast End-to-End Embedding Learning for Video Object Segmentation","arxiv_id":"1902.09513","date":"2019-02-25","proceeding":"CVPR 2019 6","authors":["Paul Voigtlaender","Yuning Chai","Florian Schroff","Hartwig Adam","Bastian Leibe","Liang-Chieh Chen"],"abstract":"Many of the recent successful methods for video object segmentation (VOS) are\noverly complicated, heavily rely on fine-tuning on the first frame, and/or are\nslow, and are hence of limited practical use. In this work, we propose FEELVOS\nas a simple and fast method which does not rely on fine-tuning. In order to\nsegment a video, for each frame FEELVOS uses a semantic pixel-wise embedding\ntogether with a global and a local matching mechanism to transfer information\nfrom the first frame and from the previous frame of the video to the current\nframe. In contrast to previous work, our embedding is only used as an internal\nguidance of a convolutional network. Our novel dynamic segmentation head allows\nus to train the network, including the embedding, end-to-end for the multiple\nobject segmentation task with a cross entropy loss. We achieve a new state of\nthe art in video object segmentation without fine-tuning with a J&F measure of\n71.5% on the DAVIS 2017 validation set. We make our code and models available\nat https://github.com/tensorflow/models/tree/master/research/feelvos.","url_abs":"http://arxiv.org/abs/1902.09513v2","url_pdf":"http://arxiv.org/pdf/1902.09513v2.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":"feelvos-fast-end-to-end-embedding-learning","repo_url":"https://github.com/tensorflow/models","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"feelvos-fast-end-to-end-embedding-learning","repo_url":"https://github.com/kim-younghan/FEELVOS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"feelvos-fast-end-to-end-embedding-learning","repo_url":"https://github.com/tensorflow/models/tree/master/research/feelvos","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"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"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-20","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS (no YouTube-VOS training)","model":"FEELVOS","rank_in_archive_order":19,"of":26,"metrics":{"D16 val (F)":"83.1","D16 val (G)":"81.7","D16 val (J)":"80.3","D17 test (F)":"57.5","D17 test (G)":"54.4","D17 test (J)":"51.2","D17 val (F)":"72.3","D17 val (G)":"69.1","D17 val (J)":"65.9","FPS":"2.22"},"uses_additional_data":false},{"leaderboard":"/sota/visual-object-tracking-on-davis-2016","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2016","model":"FEELVOS","rank_in_archive_order":59,"of":78,"metrics":{"F-measure (Decay)":"14.1","F-measure (Mean)":"82.2","F-measure (Recall)":"86.6","J&F":"81.65","Jaccard (Decay)":"13.7","Jaccard (Mean)":"81.1","Jaccard (Recall)":"90.5"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-17","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017","model":"FEELVOS","rank_in_archive_order":1,"of":1,"metrics":{"F-measure (Decay)":"20.1","F-measure (Mean)":"74.0","F-measure (Recall)":"83.8","J&F":"71.55","Jaccard (Decay)":"17.5","Jaccard (Mean)":"69.1","Jaccard (Recall)":"79.1"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-video-object-segmentation-on-1","task":"Semi-Supervised Video Object Segmentation","dataset":"DAVIS 2017 (test-dev)","model":"FEELVOS","rank_in_archive_order":47,"of":59,"metrics":{"F-measure (Decay)":"33.5","F-measure (Mean)":"60.9","F-measure (Recall)":"68.5","J&F":"57.8","Jaccard (Decay)":"29.8","Jaccard (Mean)":"55.1","Jaccard (Recall)":"62.6"},"uses_additional_data":false},{"leaderboard":"/sota/video-object-segmentation-on-youtube-1","task":"Video Object Segmentation","dataset":"YouTube","model":"FEELVOS","rank_in_archive_order":2,"of":2,"metrics":{"mIoU":"0.821"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1902.09513","atlas_url":"https://app.syntology.ai/?focus=1902.09513","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1902.09513"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/tensorflow/models","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kim-younghan/FEELVOS","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/tensorflow/models/tree/master/research/feelvos","reach":null}],"summary":{"ran_draft_wrong":1,"ran":8},"by_repo_kind":{"listed":{"samples":9,"ran":9,"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":"8c6b999f5a291101","entry":"to_np","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/trainer.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/trainer.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":2,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8c6b999f5a291101"}},{"code_sha256_prefix":"9f650f21e76de396","entry":"dice_coeff","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/metric.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/metric.py","link_basis":"harvester_set","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":"9f650f21e76de396"}},{"code_sha256_prefix":"c75bb42af05410f4","entry":"dice_loss","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/loss.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/loss.py","link_basis":"harvester_set","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":"c75bb42af05410f4"}},{"code_sha256_prefix":"1f5afcb309f3974f","entry":"distance","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/models/Matching.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/models/Matching.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1f5afcb309f3974f"}},{"code_sha256_prefix":"ed9f3272fe331772","entry":"global_matching","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/models/Matching.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/models/Matching.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ed9f3272fe331772"}},{"code_sha256_prefix":"5d2a1f4bcc55d2b8","entry":"list_to_tensor","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/util/toTensor.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/util/toTensor.py","link_basis":"harvester_set","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":"5d2a1f4bcc55d2b8"}},{"code_sha256_prefix":"ba330014042ff960","entry":"local_matching","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/models/Matching.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/models/Matching.py","link_basis":"harvester_set","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":"ba330014042ff960"}},{"code_sha256_prefix":"b9feaa87aa772f25","entry":"tversky","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/metric.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/metric.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b9feaa87aa772f25"}},{"code_sha256_prefix":"5c02e119369d3b15","entry":"tversky_loss","repo":"kim-younghan/FEELVOS","repo_kind":"listed","path":"feelvos/loss.py","file_url":"https://github.com/kim-younghan/FEELVOS/blob/HEAD/feelvos/loss.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5c02e119369d3b15"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}