{"url":"/sota/scene-flow-estimation-on-argoverse-2","task":{"name":"Scene Flow Estimation","url":"/task/scene-flow-estimation","note":null},"dataset":{"name":"Argoverse 2","url":"/dataset/argoverse-2"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Optical flow is a two-dimensional motion field in the image plane. It is the projection of the three-dimensional motion of the world. If the world is completely non-rigid, the motions of the points in the scene may all be indepen- dent of each other. One representation of the scene motion is therefore a dense **three-dimensional vector field** defined for every point on every surface in the scene. By analogy with optical flow, we refer to this three-dimensional motion field as **scene flow**.\r\n\r\nSource: Vedula, Sundar, et al. \"Three-dimensional scene flow.\" IEEE transactions on pattern analysis and machine intelligence 27.3 (2005): 475-480. [pdf](https://ieeexplore.ieee.org/document/1388274)","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["EPE 3-Way","EPE Foreground Dynamic","EPE Foreground Static","EPE Background Static"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"EPE 3-Way":null,"EPE Foreground Dynamic":null,"EPE Foreground Static":null,"EPE Background Static":null}},"counts":{"rows":7,"rows_with_code":6,"rows_with_paper_page":6,"rows_dated":6,"rows_using_additional_data":1},"rows":[{"rank_in_archive_order":1,"model":"DeFlow","metrics":{"EPE 3-Way":"0.034295","EPE Background Static":"0.004561","EPE Foreground Dynamic":"0.073231","EPE Foreground Static":"0.025093"},"uses_additional_data":false,"paper_date":"2024-01-29","paper":"/paper/deflow-decoder-of-scene-flow-network-in","paper_url":"https://arxiv.org/abs/2401.16122v1","paper_title":"DeFlow: Decoder of Scene Flow Network in Autonomous Driving","code":"https://github.com/kth-rpl/dynamicmap_benchmark","n_code_links":4,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"TrackFlow","metrics":{"EPE 3-Way":"0.047316","EPE Background Static":"0.002428","EPE Foreground Dynamic":"0.102975","EPE Foreground Static":"0.036544"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"SeFlow","metrics":{"EPE 3-Way":"0.048590","EPE Background Static":"0.005990","EPE Foreground Dynamic":"0.121375","EPE Foreground Static":"0.018404"},"uses_additional_data":false,"paper_date":"2024-07-01","paper":"/paper/seflow-a-self-supervised-scene-flow-method-in","paper_url":"https://arxiv.org/abs/2407.01702v2","paper_title":"SeFlow: A Self-Supervised Scene Flow Method in Autonomous Driving","code":"https://github.com/kth-rpl/deflow","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"ZeroFlow 5x XL","metrics":{"EPE 3-Way":"0.049392","EPE Background Static":"0.013082","EPE Foreground Dynamic":"0.117688","EPE Foreground Static":"0.017406"},"uses_additional_data":true,"paper_date":"2023-05-17","paper":"/paper/zeroflow-fast-zero-label-scene-flow-via","paper_url":"https://arxiv.org/abs/2305.10424v8","paper_title":"ZeroFlow: Scalable Scene Flow via Distillation","code":"https://github.com/kylevedder/zeroflow","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"NSFP","metrics":{"EPE 3-Way":"0.060606","EPE Background Static":"0.034445","EPE Foreground Dynamic":"0.115796","EPE Foreground Static":"0.031576"},"uses_additional_data":false,"paper_date":"2021-11-01","paper":"/paper/neural-scene-flow-prior","paper_url":"https://arxiv.org/abs/2111.01253v1","paper_title":"Neural Scene Flow Prior","code":"https://github.com/lilac-lee/neural_scene_flow_prior","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"FastFlow3D","metrics":{"EPE 3-Way":"0.061960","EPE Background Static":"0.004939","EPE Foreground Dynamic":"0.156392","EPE Foreground Static":"0.024549"},"uses_additional_data":false,"paper_date":"2021-03-01","paper":"/paper/scalable-scene-flow-from-point-clouds-in-the","paper_url":"https://arxiv.org/abs/2103.01306v5","paper_title":"Scalable Scene Flow from Point Clouds in the Real World","code":"https://github.com/kth-rpl/deflow","n_code_links":5,"syntology":null},{"rank_in_archive_order":7,"model":"FastNSF","metrics":{"EPE 3-Way":"0.111820","EPE Background Static":"0.090712","EPE Foreground Dynamic":"0.163388","EPE Foreground Static":"0.081360"},"uses_additional_data":false,"paper_date":"2023-04-18","paper":"/paper/fast-neural-scene-flow","paper_url":"https://arxiv.org/abs/2304.09121v3","paper_title":"Fast Neural Scene Flow","code":"https://github.com/lilac-lee/fastnsf","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":0}}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,885 of the 9,623 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":5,"rows_with_any_sample_ran":3,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":3,"samples_over_distinct_papers":{"n_ran":4,"n_unverified":8,"n_samples":12,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":4,"n_unverified":8,"n_samples":12,"n_pointer_only_licence":0,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}