{"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/atom-accurate-tracking-by-overlap","title":"ATOM: Accurate Tracking by Overlap Maximization","arxiv_id":"1811.07628","date":"2018-11-19","proceeding":"CVPR 2019 6","authors":["Martin Danelljan","Goutam Bhat","Fahad Shahbaz Khan","Michael Felsberg"],"abstract":"While recent years have witnessed astonishing improvements in visual tracking\nrobustness, the advancements in tracking accuracy have been limited. As the\nfocus has been directed towards the development of powerful classifiers, the\nproblem of accurate target state estimation has been largely overlooked. In\nfact, most trackers resort to a simple multi-scale search in order to estimate\nthe target bounding box. We argue that this approach is fundamentally limited\nsince target estimation is a complex task, requiring high-level knowledge about\nthe object.\n  We address this problem by proposing a novel tracking architecture,\nconsisting of dedicated target estimation and classification components. High\nlevel knowledge is incorporated into the target estimation through extensive\noffline learning. Our target estimation component is trained to predict the\noverlap between the target object and an estimated bounding box. By carefully\nintegrating target-specific information, our approach achieves previously\nunseen bounding box accuracy. We further introduce a classification component\nthat is trained online to guarantee high discriminative power in the presence\nof distractors. Our final tracking framework sets a new state-of-the-art on\nfive challenging benchmarks. On the new large-scale TrackingNet dataset, our\ntracker ATOM achieves a relative gain of 15% over the previous best approach,\nwhile running at over 30 FPS. Code and models are available at\nhttps://github.com/visionml/pytracking.","url_abs":"http://arxiv.org/abs/1811.07628v2","url_pdf":"http://arxiv.org/pdf/1811.07628v2.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":"atom-accurate-tracking-by-overlap","repo_url":"https://github.com/visionml/pytracking","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"atom-accurate-tracking-by-overlap","repo_url":"https://github.com/martin-danelljan/Continuous-ConvOp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"atom-accurate-tracking-by-overlap","repo_url":"https://github.com/martin-danelljan/ECO","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"atom-accurate-tracking-by-overlap","repo_url":"https://github.com/xuefeng-zhu5/cdaat","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"state-estimation","task_name":"State Estimation"},{"task_slug":"visual-object-tracking","task_name":"Visual Object Tracking"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/object-tracking-on-fe108","task":"Object Tracking","dataset":"FE108","model":"ATOM","rank_in_archive_order":7,"of":8,"metrics":{"Averaged Precision":"71.3","Success Rate":"46.5"},"uses_additional_data":false},{"leaderboard":"/sota/visual-object-tracking-on-got-10k","task":"Visual Object Tracking","dataset":"GOT-10k","model":"ATOM","rank_in_archive_order":41,"of":42,"metrics":{"Average Overlap":"61.0","Success Rate 0.5":"74.2"},"uses_additional_data":false},{"leaderboard":"/sota/visual-object-tracking-on-lasot","task":"Visual Object Tracking","dataset":"LaSOT","model":"ATOM","rank_in_archive_order":44,"of":46,"metrics":{"AUC":"51.4","Normalized Precision":"57.6","Precision":"50.5"},"uses_additional_data":false},{"leaderboard":"/sota/visual-object-tracking-on-trackingnet","task":"Visual Object Tracking","dataset":"TrackingNet","model":"ATOM","rank_in_archive_order":33,"of":40,"metrics":{"Accuracy":"70.34","Normalized Precision":"77.11","Precision":"64.84"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.07628","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1811.07628"}},"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/xuefeng-zhu5/cdaat","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/visionml/pytracking","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/martin-danelljan/ECO","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/martin-danelljan/Continuous-ConvOp","reach":{"status":"ok","spdx":"GPL-3.0"}}],"summary":{"ran_honours":1,"ran":2,"unverified":6},"by_repo_kind":{"listed":{"samples":9,"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":"e0a06ded5d4f6c3c","entry":"box_cxcywh_to_xyxy","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/utils/box_ops.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/utils/box_ops.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e0a06ded5d4f6c3c"}},{"code_sha256_prefix":"40c655a7fe309e49","entry":"box_xywh_to_xyxy","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/utils/box_ops.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/utils/box_ops.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":"40c655a7fe309e49"}},{"code_sha256_prefix":"b48a27604b9574e0","entry":"box_xyxy_to_xywh","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/utils/box_ops.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/utils/box_ops.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":"b48a27604b9574e0"}},{"code_sha256_prefix":"ce98e7db9de27859","entry":"all_gather","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/utils/misc.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/utils/misc.py","link_basis":"harvester_set","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":"ce98e7db9de27859"}},{"code_sha256_prefix":"8e3a5839c7c54607","entry":"collate_fn","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/utils/misc.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/utils/misc.py","link_basis":"harvester_set","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":"8e3a5839c7c54607"}},{"code_sha256_prefix":"e857124b392e7d82","entry":"get_optimizer_scheduler","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/train/base_functions.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/train/base_functions.py","link_basis":"harvester_set","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":"e857124b392e7d82"}},{"code_sha256_prefix":"604be06d5f714be8","entry":"merge_feature_sequence","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/utils/merge.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/utils/merge.py","link_basis":"harvester_set","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":"604be06d5f714be8"}},{"code_sha256_prefix":"5c8f207cb5f19197","entry":"reduce_dict","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/utils/misc.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/utils/misc.py","link_basis":"harvester_set","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":"5c8f207cb5f19197"}},{"code_sha256_prefix":"2c47f808b8359c8f","entry":"tensor_operation","repo":"xuefeng-zhu5/cdaat","repo_kind":"listed","path":"lib/utils/tensor.py","file_url":"https://github.com/xuefeng-zhu5/cdaat/blob/HEAD/lib/utils/tensor.py","link_basis":"harvester_set","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":"2c47f808b8359c8f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}