{"url":"/sota/image-matching-on-zeb","task":{"name":"Image Matching","url":"/task/image-matching","note":null},"dataset":{"name":"ZEB","url":"/dataset/zeb"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":null,"description_from":null,"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":["Mean AUC@5°"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Mean AUC@5°":"higher"}},"counts":{"rows":10,"rows_with_code":10,"rows_with_paper_page":10,"rows_dated":10,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"GIM-RoMa","metrics":{"Mean AUC@5°":"53.3"},"uses_additional_data":false,"paper_date":"2024-02-16","paper":"/paper/gim-learning-generalizable-image-matcher-from","paper_url":"https://arxiv.org/abs/2402.11095v1","paper_title":"GIM: Learning Generalizable Image Matcher From Internet Videos","code":"https://github.com/xuelunshen/gim","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":5,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"GIM-DKM","metrics":{"Mean AUC@5°":"51.2"},"uses_additional_data":false,"paper_date":"2024-02-16","paper":"/paper/gim-learning-generalizable-image-matcher-from","paper_url":"https://arxiv.org/abs/2402.11095v1","paper_title":"GIM: Learning Generalizable Image Matcher From Internet Videos","code":"https://github.com/xuelunshen/gim","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":5,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"RoMa","metrics":{"Mean AUC@5°":"48.8"},"uses_additional_data":false,"paper_date":"2023-05-24","paper":"/paper/roma-revisiting-robust-losses-for-dense","paper_url":"https://arxiv.org/abs/2305.15404v2","paper_title":"RoMa: Robust Dense Feature Matching","code":"https://github.com/parskatt/roma","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"DKM","metrics":{"Mean AUC@5°":"46.2"},"uses_additional_data":false,"paper_date":"2022-02-01","paper":"/paper/deep-kernelized-dense-geometric-matching","paper_url":"https://arxiv.org/abs/2202.00667v3","paper_title":"DKM: Dense Kernelized Feature Matching for Geometry Estimation","code":"https://github.com/parskatt/dkm","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"GIM-LoFTR","metrics":{"Mean AUC@5°":"39.1"},"uses_additional_data":false,"paper_date":"2024-02-16","paper":"/paper/gim-learning-generalizable-image-matcher-from","paper_url":"https://arxiv.org/abs/2402.11095v1","paper_title":"GIM: Learning Generalizable Image Matcher From Internet Videos","code":"https://github.com/xuelunshen/gim","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":5,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"GIM-LightGlue","metrics":{"Mean AUC@5°":"38.3"},"uses_additional_data":false,"paper_date":"2024-02-16","paper":"/paper/gim-learning-generalizable-image-matcher-from","paper_url":"https://arxiv.org/abs/2402.11095v1","paper_title":"GIM: Learning Generalizable Image Matcher From Internet Videos","code":"https://github.com/xuelunshen/gim","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":5,"n_samples":17,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"LoFTR","metrics":{"Mean AUC@5°":"33.1"},"uses_additional_data":false,"paper_date":"2021-04-01","paper":"/paper/loftr-detector-free-local-feature-matching","paper_url":"https://arxiv.org/abs/2104.00680v1","paper_title":"LoFTR: Detector-Free Local Feature Matching with Transformers","code":"https://github.com/kornia/kornia","n_code_links":4,"syntology":{"n_ran":14,"n_unverified":2,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"RootSIFT","metrics":{"Mean AUC@5°":"31.8"},"uses_additional_data":false,"paper_date":"2012-06-16","paper":"/paper/three-things-everyone-should-know-to-improve","paper_url":"https://ieeexplore.ieee.org/document/6248018","paper_title":"Three things everyone should know to improve object retrieval","code":"https://github.com/ubc-vision/image-matching-benchmark","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"LightGlue","metrics":{"Mean AUC@5°":"31.7"},"uses_additional_data":false,"paper_date":"2023-06-23","paper":"/paper/lightglue-local-feature-matching-at-light","paper_url":"https://arxiv.org/abs/2306.13643v1","paper_title":"LightGlue: Local Feature Matching at Light Speed","code":"https://github.com/cvg/lightglue","n_code_links":2,"syntology":{"n_ran":18,"n_unverified":22,"n_samples":40,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"SuperGlue","metrics":{"Mean AUC@5°":"31.2"},"uses_additional_data":false,"paper_date":"2019-11-26","paper":"/paper/superglue-learning-feature-matching-with","paper_url":"https://arxiv.org/abs/1911.11763v2","paper_title":"SuperGlue: Learning Feature Matching with Graph Neural Networks","code":"https://github.com/huggingface/transformers","n_code_links":19,"syntology":{"n_ran":6,"n_unverified":16,"n_samples":22,"n_pointer_only_licence":6}}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"syntology":{"read_at":"2026-09-24T18:15:14+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":8,"rows_with_any_sample_ran":7,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":50,"n_unverified":47,"n_samples":97,"n_pointer_only_licence":6,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":86,"n_unverified":62,"n_samples":148,"n_pointer_only_licence":6,"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"}}}