{"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/tapir-tracking-any-point-with-per-frame","title":"TAPIR: Tracking Any Point with per-frame Initialization and temporal Refinement","arxiv_id":"2306.08637","date":"2023-06-14","proceeding":"ICCV 2023 1","authors":["Carl Doersch","Yi Yang","Mel Vecerik","Dilara Gokay","Ankush Gupta","Yusuf Aytar","Joao Carreira","Andrew Zisserman"],"abstract":"We present a novel model for Tracking Any Point (TAP) that effectively tracks any queried point on any physical surface throughout a video sequence. Our approach employs two stages: (1) a matching stage, which independently locates a suitable candidate point match for the query point on every other frame, and (2) a refinement stage, which updates both the trajectory and query features based on local correlations. The resulting model surpasses all baseline methods by a significant margin on the TAP-Vid benchmark, as demonstrated by an approximate 20% absolute average Jaccard (AJ) improvement on DAVIS. Our model facilitates fast inference on long and high-resolution video sequences. On a modern GPU, our implementation has the capacity to track points faster than real-time, and can be flexibly extended to higher-resolution videos. Given the high-quality trajectories extracted from a large dataset, we demonstrate a proof-of-concept diffusion model which generates trajectories from static images, enabling plausible animations. Visualizations, source code, and pretrained models can be found on our project webpage.","url_abs":"https://arxiv.org/abs/2306.08637v2","url_pdf":"https://arxiv.org/pdf/2306.08637v2.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":"tapir-tracking-any-point-with-per-frame","repo_url":"https://github.com/deepmind/tapnet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"jax","reach":null},{"paper_slug":"tapir-tracking-any-point-with-per-frame","repo_url":"https://github.com/ibaiGorordo/Tapir-Pytorch-Inference","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"tapir-tracking-any-point-with-per-frame","repo_url":"https://github.com/riponazad/echotracker","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":null,"task_name":"GPU"},{"task_slug":"motion-estimation","task_name":"Motion Estimation"},{"task_slug":"point-tracking","task_name":"Point Tracking"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"}],"methods":[{"method_slug":"depthwise-convolution","method_name":"Depthwise Convolution"},{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/visual-tracking-on-davis","task":"Visual Tracking","dataset":"DAVIS","model":"TAPIR (Panning MOVi-E)","rank_in_archive_order":1,"of":2,"metrics":{"Average Jaccard":"61.3"},"uses_additional_data":false},{"leaderboard":"/sota/visual-tracking-on-davis","task":"Visual Tracking","dataset":"DAVIS","model":"TAPIR (MOVi-E)","rank_in_archive_order":2,"of":2,"metrics":{"Average Jaccard":"59.8"},"uses_additional_data":false},{"leaderboard":"/sota/visual-tracking-on-kinetics","task":"Visual Tracking","dataset":"Kinetics","model":"TAPIR (Panning MOVi-E)","rank_in_archive_order":1,"of":2,"metrics":{"Average Jaccard":"57.2"},"uses_additional_data":false},{"leaderboard":"/sota/visual-tracking-on-kinetics","task":"Visual Tracking","dataset":"Kinetics","model":"TAPIR (MOVi-E)","rank_in_archive_order":2,"of":2,"metrics":{"Average Jaccard":"57.1"},"uses_additional_data":false},{"leaderboard":"/sota/visual-tracking-on-kubric","task":"Visual Tracking","dataset":"Kubric","model":"TAPIR (Panning MOVi-E)","rank_in_archive_order":1,"of":2,"metrics":{"Average Jaccard":"84.7"},"uses_additional_data":false},{"leaderboard":"/sota/visual-tracking-on-kubric","task":"Visual Tracking","dataset":"Kubric","model":"TAPIR (MOVi-E)","rank_in_archive_order":2,"of":2,"metrics":{"Average Jaccard":"84.3"},"uses_additional_data":false},{"leaderboard":"/sota/visual-tracking-on-rgb-stacking","task":"Visual Tracking","dataset":"RGB-Stacking","model":"TAPIR (MOVi-E)","rank_in_archive_order":1,"of":2,"metrics":{"Average Jaccard":"66.2"},"uses_additional_data":false},{"leaderboard":"/sota/visual-tracking-on-rgb-stacking","task":"Visual Tracking","dataset":"RGB-Stacking","model":"TAPIR (Panning MOVi-E)","rank_in_archive_order":2,"of":2,"metrics":{"Average Jaccard":"62.7"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2306.08637","atlas_url":"https://app.syntology.ai/?focus=2306.08637","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2306.08637"}},"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":"deterministic:regex_extraction","url":"https://github.com/deepmind/tapnet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/riponazad/echotracker","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ibaiGorordo/Tapir-Pytorch-Inference","reach":null}],"summary":{"unverified":1},"by_repo_kind":{"listed":{"samples":1,"ran":0,"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":"bc0147f0be80b100","entry":"TAPIR","repo":"ibaiGorordo/Tapir-Pytorch-Inference","repo_kind":"listed","path":"tapnet/tapir_model.py","file_url":"https://github.com/ibaiGorordo/Tapir-Pytorch-Inference/blob/HEAD/tapnet/tapir_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bc0147f0be80b100"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}