{"url":"/dataset/faust-partial-1","name":"FAUST-partial","full_name":null,"description_markdown":"FAUST-partial is a 3D registration benchmark dataset created to provide a more informative evaluation of 3D registration methods. \r\nThe dataset addresses two main limitations of current 3D registration benchmarks:\r\n\r\n1. Lack of data variability within the current registration benchmarks\r\n2. Capability to evaluate the registration method on a single 3D registration parameter (rotation, translation or overlap)\r\n\r\nThe benchmark is created using the FAUST training dataset comprised of 100 3D scans of human bodies. Note however, that the methodology can be applied to any point cloud dataset.\r\n\r\nThe benchmark generation for a single scan from the FAUST training dataset can be summarized as follows:\r\n\r\n- Surround the scan with a regular icosahaedron. Each point of the icosahaedron acts as a viewpoint\r\n- For each viewpoint, create a partial point cloud using the hidden point removal algorithm\r\n- Finally, for a pair of partial point clouds in a desired range of overlap, generate a random rotation from the desired rotation and translation ranges.\r\n\r\nTo create a more informative 3D registration benchmark, 3 difficulty settings are used: Easy (E), Medium (M) and Hard (H) for each registration parameter: Rotation (R), Translation (T) and Overlap (O). This way, 9 benchmarks are created in which each triplet increases the difficulty of one registration parameter whilst fixing the difficulty of the other two. This allows to evaluate a 3D registration method w.r.t. only one registration parameter and observe the robustness of the method w.r.t. that parameter.\r\n\r\nThe benchmarks are denominated as FP-{R,T,O}-{E,M,H}.","description_withheld":null,"homepage":"https://github.com/DavidBoja/exhaustive-grid-search","introduced_date":"2024-03-23","introduced_date_note":null,"introduced_by":{"paper":"/paper/addressing-the-generalization-of-3d","title":"Addressing the generalization of 3D registration methods with a featureless baseline and an unbiased benchmark","first_author":"Bojanić","url":null},"license":null,"modalities":[{"name":"3D","url":"/datasets/modality/3d"}],"tasks":[{"name":"Point Cloud Registration","url":"/task/point-cloud-registration","datasets_with_task":"/datasets/task/point-cloud-registration"}],"languages":[],"variants":["FP-O-H","FP-O-M","FP-O-E","FP-T-H","FP-T-M","FP-T-E","FP-R-H","FP-R-M","FP-R-E","FAUST-partial"],"data_loaders":[],"num_papers_in_archive":5,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/point-cloud-registration-on-fp-o-e","task":"Point Cloud Registration","dataset_variant":"FP-O-E","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"FPFH+SC2-PCR","paper":"/paper/sc2-pcr-a-second-order-spatial-compatibility","metrics":{"RRE (degrees)":"0.91","RTE (cm)":"0.43","Recall (3cm, 10 degrees)":"99.88"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/point-cloud-registration-on-fp-o-h","task":"Point Cloud Registration","dataset_variant":"FP-O-H","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"FPFH+SC2-PCR","paper":"/paper/sc2-pcr-a-second-order-spatial-compatibility","metrics":{"RRE (degrees)":"2.42","RTE (cm)":"1.23","Recall (3cm, 10 degrees)":"38.85"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/point-cloud-registration-on-fp-o-m","task":"Point Cloud Registration","dataset_variant":"FP-O-M","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"Exhaustive Grid Search","paper":"/paper/addressing-the-generalization-of-3d","metrics":{"RRE (degrees)":"0.030","RTE (cm)":"0.017","Recall (3cm, 10 degrees)":"88.06"},"code_links":[{"title":"DavidBoja/exhaustive-grid-search","url":"https://github.com/DavidBoja/exhaustive-grid-search"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/point-cloud-registration-on-fp-r-e","task":"Point Cloud Registration","dataset_variant":"FP-R-E","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"GeDi","paper":"/paper/generalisable-and-distinctive-3d-local-deep","metrics":{"RRE (degrees)":"1.629","RTE (cm)":"1.162","Recall (3cm, 10 degrees)":"99.76"},"code_links":[{"title":"fabiopoiesi/gedi","url":"https://github.com/fabiopoiesi/gedi"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/point-cloud-registration-on-fp-r-h","task":"Point Cloud Registration","dataset_variant":"FP-R-H","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"GeDi","paper":"/paper/generalisable-and-distinctive-3d-local-deep","metrics":{"RRE (degrees)":"1.70","RTE (cm)":"1.63","Recall (3cm, 10 degrees)":"99.41"},"code_links":[{"title":"fabiopoiesi/gedi","url":"https://github.com/fabiopoiesi/gedi"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/point-cloud-registration-on-fp-r-m","task":"Point Cloud Registration","dataset_variant":"FP-R-M","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"GeDi","paper":"/paper/generalisable-and-distinctive-3d-local-deep","metrics":{"RRE (degrees)":"1.66","RTE (cm)":"1.14","Recall (3cm, 10 degrees)":"99.94"},"code_links":[{"title":"fabiopoiesi/gedi","url":"https://github.com/fabiopoiesi/gedi"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/point-cloud-registration-on-fp-t-e","task":"Point Cloud Registration","dataset_variant":"FP-T-E","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"FPFH+SC2-PCR","paper":"/paper/sc2-pcr-a-second-order-spatial-compatibility","metrics":{"RRE (degrees)":"0.93","RTE (cm)":"0.43","Recall (3cm, 10 degrees)":"99.76"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/point-cloud-registration-on-fp-t-h","task":"Point Cloud Registration","dataset_variant":"FP-T-H","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"GeDi","paper":"/paper/generalisable-and-distinctive-3d-local-deep","metrics":{"RRE (degrees)":"1.63","RTE (cm)":"1.14","Recall (3cm, 10 degrees)":"99.70"},"code_links":[{"title":"fabiopoiesi/gedi","url":"https://github.com/fabiopoiesi/gedi"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/point-cloud-registration-on-fp-t-m","task":"Point Cloud Registration","dataset_variant":"FP-T-M","rows":6,"metrics":["Recall (3cm, 10 degrees)","RRE (degrees)","RTE (cm)"],"first_row_in_archive_order":{"model":"Exhaustive Grid Search","paper":"/paper/addressing-the-generalization-of-3d","metrics":{"RRE (degrees)":"0.005","RTE (cm)":"0.002","Recall (3cm, 10 degrees)":"99.82"},"code_links":[{"title":"DavidBoja/exhaustive-grid-search","url":"https://github.com/DavidBoja/exhaustive-grid-search"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/addressing-the-generalization-of-3d","title":"Addressing the generalization of 3D registration methods with a featureless baseline and an unbiased benchmark","date":"2024-03-23","rows_on_this_dataset":9,"code_links":1,"syntology":null},{"paper":"/paper/geotransformer-fast-and-robust-point-cloud","title":"GeoTransformer: Fast and Robust Point Cloud Registration with Geometric Transformer","date":"2023-07-25","rows_on_this_dataset":9,"code_links":1,"syntology":null},{"paper":"/paper/sc2-pcr-a-second-order-spatial-compatibility","title":"SC2-PCR: A Second Order Spatial Compatibility for Efficient and Robust Point Cloud Registration","date":"2022-01-01","rows_on_this_dataset":18,"code_links":0,"syntology":null},{"paper":"/paper/generalisable-and-distinctive-3d-local-deep","title":"Learning general and distinctive 3D local deep descriptors for point cloud registration","date":"2021-05-21","rows_on_this_dataset":9,"code_links":1,"syntology":null},{"paper":"/paper/go-icp-a-globally-optimal-solution-to-3d-icp","title":"Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registration","date":"2016-05-11","rows_on_this_dataset":9,"code_links":0,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}