{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/geotransformer-fast-and-robust-point-cloud","title":"GeoTransformer: Fast and Robust Point Cloud Registration with Geometric Transformer","arxiv_id":"2308.03768","date":"2023-07-25","proceeding":null,"authors":["Zheng Qin","Hao Yu","Changjian Wang","Yulan Guo","Yuxing Peng","Slobodan Ilic","Dewen Hu","Kai Xu"],"abstract":"We study the problem of extracting accurate correspondences for point cloud registration. Recent keypoint-free methods have shown great potential through bypassing the detection of repeatable keypoints which is difficult to do especially in low-overlap scenarios. They seek correspondences over downsampled superpoints, which are then propagated to dense points. Superpoints are matched based on whether their neighboring patches overlap. Such sparse and loose matching requires contextual features capturing the geometric structure of the point clouds. We propose Geometric Transformer, or GeoTransformer for short, to learn geometric feature for robust superpoint matching. It encodes pair-wise distances and triplet-wise angles, making it invariant to rigid transformation and robust in low-overlap cases. The simplistic design attains surprisingly high matching accuracy such that no RANSAC is required in the estimation of alignment transformation, leading to $100$ times acceleration. Extensive experiments on rich benchmarks encompassing indoor, outdoor, synthetic, multiway and non-rigid demonstrate the efficacy of GeoTransformer. Notably, our method improves the inlier ratio by $18{\\sim}31$ percentage points and the registration recall by over $7$ points on the challenging 3DLoMatch benchmark. Our code and models are available at \\url{https://github.com/qinzheng93/GeoTransformer}.","url_abs":"https://arxiv.org/abs/2308.03768v1","url_pdf":"https://arxiv.org/pdf/2308.03768v1.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":"geotransformer-fast-and-robust-point-cloud","repo_url":"https://github.com/qinzheng93/geotransformer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-to-point-cloud-registration","task_name":"Image to Point Cloud Registration"},{"task_slug":"point-cloud-registration","task_name":"Point Cloud Registration"},{"task_slug":null,"task_name":"Triplet"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/point-cloud-registration-on-eth-trained-on","task":"Point Cloud Registration","dataset":"ETH (trained on 3DMatch)","model":"GeoTransformer","rank_in_archive_order":19,"of":20,"metrics":{"Recall (30cm, 5 degrees)":"4.91"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-o-e","task":"Point Cloud Registration","dataset":"FP-O-E","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"2.30","RTE (cm)":"1.57","Recall (3cm, 10 degrees)":"63.94"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-o-h","task":"Point Cloud Registration","dataset":"FP-O-H","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"2.57","RTE (cm)":"2.22","Recall (3cm, 10 degrees)":"2.64"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-o-m","task":"Point Cloud Registration","dataset":"FP-O-M","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"2.45","RTE (cm)":"1.94","Recall (3cm, 10 degrees)":"22.07"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-r-e","task":"Point Cloud Registration","dataset":"FP-R-E","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"2.29","RTE (cm)":"1.57","Recall (3cm, 10 degrees)":"64.12"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-r-h","task":"Point Cloud Registration","dataset":"FP-R-H","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"0.47","RTE (cm)":"1.69","Recall (3cm, 10 degrees)":"47.75"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-r-m","task":"Point Cloud Registration","dataset":"FP-R-M","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"2.26","RTE (cm)":"1.63","Recall (3cm, 10 degrees)":"55.93"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-t-e","task":"Point Cloud Registration","dataset":"FP-T-E","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"2.32","RTE (cm)":"1.59","Recall (3cm, 10 degrees)":"66.25"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-t-h","task":"Point Cloud Registration","dataset":"FP-T-H","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"2.27","RTE (cm)":"1.57","Recall (3cm, 10 degrees)":"64.18"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-fp-t-m","task":"Point Cloud Registration","dataset":"FP-T-M","model":"GeoTransformer","rank_in_archive_order":5,"of":6,"metrics":{"RRE (degrees)":"2.29","RTE (cm)":"1.58","Recall (3cm, 10 degrees)":"64.29"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-kitti-trained-on","task":"Point Cloud Registration","dataset":"KITTI (trained on 3DMatch)","model":"GeoTransformer","rank_in_archive_order":11,"of":14,"metrics":{"Success Rate":"67.93"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-registration-on-rotkitti","task":"Point Cloud Registration","dataset":"RotKITTI Registration Benchmark","model":"GeoTransformer","rank_in_archive_order":2,"of":6,"metrics":{"RR@(1,0.1)":"50.1","RR@(1.5,0.3)":"78.5"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2308.03768","atlas_url":"https://app.syntology.ai/?focus=2308.03768","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2308.03768"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2308.03768\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2308.03768"},"call":"get_concepts_for_paper(arxiv_id=\"2308.03768\")"}],"by_repo":[{"repo":"qinzheng93/geotransformer","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"edecc1e2960d4256fc4ae8f823d2e90db611badf8e0b1efde36122bc45bcc043","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}