{"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/point-transformer-1","title":"Point Transformer","arxiv_id":"2012.09164","date":"2020-12-16","proceeding":"ICCV 2021 10","authors":["Hengshuang Zhao","Li Jiang","Jiaya Jia","Philip Torr","Vladlen Koltun"],"abstract":"Self-attention networks have revolutionized natural language processing and are making impressive strides in image analysis tasks such as image classification and object detection. Inspired by this success, we investigate the application of self-attention networks to 3D point cloud processing. We design self-attention layers for point clouds and use these to construct self-attention networks for tasks such as semantic scene segmentation, object part segmentation, and object classification. Our Point Transformer design improves upon prior work across domains and tasks. For example, on the challenging S3DIS dataset for large-scale semantic scene segmentation, the Point Transformer attains an mIoU of 70.4% on Area 5, outperforming the strongest prior model by 3.3 absolute percentage points and crossing the 70% mIoU threshold for the first time.","url_abs":"https://arxiv.org/abs/2012.09164v2","url_pdf":"https://arxiv.org/pdf/2012.09164v2.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":"point-transformer-1","repo_url":"https://github.com/Pointcept/Pointcept","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/2023-MindSpore-1/ms-code-48","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mindspore","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/KernelA/pytorch-point-transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/Meowuu7/Point-Transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/POSTECH-CVLab/FastPointTransformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/POSTECH-CVLab/point-transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/Sharpiless/Point-Transformer-Pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/alzmzyy/pointtransformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mindspore","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/lucidrains/point-transformer-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/qq456cvb/Point-Transformers","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/rauleun/point-transformer-tf2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/tkdguraa/point-transformer-tensorflow","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://gitee.com/gai-shaoyan/mind3d","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/2023-MindSpore-1/ms-code-215/tree/main/pointtransformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/MS-Mind/MS-Code-06/tree/main/pointtransformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/Mind23-2/MindCode-5/tree/main/pointtransformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/MindSpore-paper-code-2/code2/tree/main/pointtransformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/code-implementation1/Code6/tree/main/pointtransformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/engelnico/point-transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/gofinge/pointtransformerv2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/isl-org/Open3D-ML","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/jxl152/Point-Transformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/lifebeyondexpectations/eccv22-pointmixer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"point-transformer-1","repo_url":"https://github.com/sh4175515/PointTransformer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"3d-part-segmentation","task_name":"3D Part Segmentation"},{"task_slug":"3d-point-cloud-classification","task_name":"3D Point Cloud Classification"},{"task_slug":"3d-semantic-segmentation","task_name":"3D Semantic Segmentation"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"point-cloud-segmentation","task_name":"Point Cloud Segmentation"},{"task_slug":"scene-segmentation","task_name":"Scene Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"image-classification","task_name":"image-classification"}],"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/3d-part-segmentation-on-shapenet-part","task":"3D Part Segmentation","dataset":"ShapeNet-Part","model":"PointTransformer","rank_in_archive_order":19,"of":67,"metrics":{"Class Average IoU":"83.7","Instance Average IoU":"86.6"},"uses_additional_data":false},{"leaderboard":"/sota/3d-point-cloud-classification-on-modelnet40","task":"3D Point Cloud Classification","dataset":"ModelNet40","model":"PointTransformer","rank_in_archive_order":50,"of":111,"metrics":{"Mean Accuracy":"90.6","Overall Accuracy":"93.7"},"uses_additional_data":false},{"leaderboard":"/sota/3d-semantic-segmentation-on-s3dis","task":"3D Semantic Segmentation","dataset":"S3DIS","model":"PointTransformer","rank_in_archive_order":4,"of":6,"metrics":{"mIoU (6-Fold)":"73.5","mIoU (Area-5)":"70.4"},"uses_additional_data":false},{"leaderboard":"/sota/3d-semantic-segmentation-on-stpls3d","task":"3D Semantic Segmentation","dataset":"STPLS3D","model":"Point transformer","rank_in_archive_order":5,"of":6,"metrics":{"mIOU":"47.64"},"uses_additional_data":false},{"leaderboard":"/sota/point-cloud-segmentation-on-pointcloud-c","task":"Point Cloud Segmentation","dataset":"PointCloud-C","model":"PointTransformers","rank_in_archive_order":7,"of":11,"metrics":{"mean Corruption Error (mCE)":"1.049"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-s3dis","task":"Semantic Segmentation","dataset":"S3DIS","model":"PointTransformer","rank_in_archive_order":17,"of":54,"metrics":{"Mean IoU":"73.5","Number of params":"7.8M","Params (M)":"7.8","mAcc":"81.9","oAcc":"90.2"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-s3dis","task":"Semantic Segmentation","dataset":"S3DIS","model":"KPConv","rank_in_archive_order":24,"of":54,"metrics":{"Mean IoU":"70.6","Number of params":"14.1M","Params (M)":"14.1"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-s3dis","task":"Semantic Segmentation","dataset":"S3DIS","model":"PointCNN","rank_in_archive_order":36,"of":54,"metrics":{"Mean IoU":"65.4","Number of params":"N/A"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-s3dis","task":"Semantic Segmentation","dataset":"S3DIS","model":"SPGraph","rank_in_archive_order":42,"of":54,"metrics":{"Mean IoU":"62.1","Number of params":"N/A"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-s3dis","task":"Semantic Segmentation","dataset":"S3DIS","model":"PointNet","rank_in_archive_order":51,"of":54,"metrics":{"Mean IoU":"47.6","Number of params":"N/A"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-s3dis-area5","task":"Semantic Segmentation","dataset":"S3DIS Area5","model":"PointTransformer","rank_in_archive_order":32,"of":61,"metrics":{"Number of params":"7.8M","mAcc":"76.5","mIoU":"70.4","oAcc":"90.8"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-s3dis-area5","task":"Semantic Segmentation","dataset":"S3DIS Area5","model":"PointCNN","rank_in_archive_order":53,"of":61,"metrics":{"Number of params":"N/A","mIoU":"57.3"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-s3dis-area5","task":"Semantic Segmentation","dataset":"S3DIS Area5","model":"PointNet","rank_in_archive_order":56,"of":61,"metrics":{"Number of params":"N/A","mIoU":"41.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2012.09164","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}