{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T10:24:44Z","timestamp":1772101484018,"version":"3.50.1"},"reference-count":56,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2023,3,20]],"date-time":"2023-03-20T00:00:00Z","timestamp":1679270400000},"content-version":"vor","delay-in-days":170,"URL":"https:\/\/2.zoppoz.workers.dev:443\/http\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62172218"],"award-info":[{"award-number":["62172218"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62032011"],"award-info":[{"award-number":["62032011"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003453","name":"Natural Science Foundation of Guangdong Province","doi-asserted-by":"publisher","award":["2022A1515010170"],"award-info":[{"award-number":["2022A1515010170"]}],"id":[{"id":"10.13039\/501100003453","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Computer Graphics Forum"],"published-print":{"date-parts":[[2022,10]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>6D pose estimation of rigid objects from RGB\u2010D images is crucial for object grasping and manipulation in robotics. Although RGB channels and the depth (D) channel are often complementary, providing respectively the appearance and geometry information, it is still non\u2010trivial on how to fully benefit from the two cross\u2010modal data. From the simple yet new observation, when an object rotates, its semantic label is invariant to the pose while its keypoint offset direction is variant to the pose. To this end, we present SO(3)\u2010Pose, a new representation learning network to explore SO(3)\u2010equivariant and SO(3)\u2010invariant features from the depth channel for pose estimation. The SO(3)\u2010invariant features facilitate to learn more distinctive representations for segmenting objects with similar appearance from RGB channels. The SO(3)\u2010equivariant features communicate with RGB features to deduce the (missed) geometry for detecting keypoints of an object with the reflective surface from the depth channel. Unlike most of existing pose estimation methods, our SO(3)\u2010Pose not only implements the information communication between the RGB and depth channels, but also naturally absorbs the SO(3)\u2010equivariance geometry knowledge from depth images, leading to better appearance and geometry representation learning. Comprehensive experiments show that our method achieves the state\u2010of\u2010the\u2010art performance on three benchmarks. Code is available at <jats:ext-link xmlns:xlink=\"https:\/\/2.zoppoz.workers.dev:443\/http\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/2.zoppoz.workers.dev:443\/https\/github.com\/phaoran9999\/SO3-Pose\">https:\/\/2.zoppoz.workers.dev:443\/https\/github.com\/phaoran9999\/SO3-Pose<\/jats:ext-link>.<\/jats:p>","DOI":"10.1111\/cgf.14684","type":"journal-article","created":{"date-parts":[[2023,3,20]],"date-time":"2023-03-20T13:46:27Z","timestamp":1679319987000},"page":"371-381","update-policy":"https:\/\/2.zoppoz.workers.dev:443\/https\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["SO(3)\u2010Pose: SO(3)\u2010Equivariance Learning for 6D Object Pose Estimation"],"prefix":"10.1111","volume":"41","author":[{"given":"Haoran","family":"Pan","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology Nanjing University of Aeronautics and Astronautics  Nanjing China"},{"name":"Shenzhen Research Institute Nanjing University of Aeronautics and Astronautics  Shenzhen China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jun","family":"Zhou","sequence":"additional","affiliation":[{"name":"School of Nursing Hong Kong Polytechnic University  Hong Kong China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuanpeng","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Nanjing University of Aeronautics and Astronautics  Nanjing China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xuequan","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Information Technology Deakin University  Geelong Australia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weiming","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Science and Technology Hong Kong Metropolitan University  Hong Kong China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xuefeng","family":"Yan","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Nanjing University of Aeronautics and Astronautics  Nanjing China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mingqiang","family":"Wei","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology Nanjing University of Aeronautics and Astronautics  Nanjing China"},{"name":"Shenzhen Research Institute Nanjing University of Aeronautics and Astronautics  Shenzhen China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2023,3,20]]},"reference":[{"key":"e_1_2_7_2_2","article-title":"Xcit: Cross-covariance image transformers","volume":"34","author":"Ali A.","year":"2021","journal-title":"Advances in neural information processing systems"},{"key":"e_1_2_7_3_2","unstructured":"BaJ. KirosJ. R. HintonG. E.: Layer normalization.arXiv: Machine Learning(2016). 5"},{"key":"e_1_2_7_4_2","doi-asserted-by":"crossref","unstructured":"BrachmannE. KrullA. MichelF. GumholdS. ShottonJ. RotherC.: Learning 6d object pose estimation using 3d object coordinates.european conference on computer vision(2014). 6","DOI":"10.1007\/978-3-319-10605-2_35"},{"key":"e_1_2_7_5_2","doi-asserted-by":"crossref","unstructured":"ChenW. JiaX. ChangH. J. DuanJ. LeonardisA.: G2l-net: Global to local network for real-time 6d pose estimation with embedding vector features.computer vision and pattern recognition(2020). 8","DOI":"10.1109\/CVPR42600.2020.00429"},{"key":"e_1_2_7_6_2","doi-asserted-by":"crossref","unstructured":"CalliB. SinghA. WalsmanA. SrinivasaS. S. AbbeelP. DollarA. M.: The ycb object and model set: Towards common benchmarks for manipulation research.international conference on advanced robotics(2015). 6","DOI":"10.1109\/ICAR.2015.7251504"},{"key":"e_1_2_7_7_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2012.6225371"},{"key":"e_1_2_7_8_2","doi-asserted-by":"crossref","unstructured":"DaiZ. CaiB. LinY. ChenJ.: Up-detr: Unsupervised pre-training for object detection with transformers. InProceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition(2021) pp.1601\u20131610. 3","DOI":"10.1109\/CVPR46437.2021.00165"},{"key":"e_1_2_7_9_2","doi-asserted-by":"crossref","unstructured":"DengC. LitanyO. DuanY. PoulenardA. TagliasacchiA. GuibasL. J.: Vector neurons: A general framework for so (3)-equivariant networks. InProceedings of the IEEE\/CVF International Conference on Computer Vision(2021) pp.12200\u201312209. 4 5 6","DOI":"10.1109\/ICCV48922.2021.01198"},{"key":"e_1_2_7_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2010.5540108"},{"key":"e_1_2_7_11_2","unstructured":"DuG. WangK. LianS.:Vision-based robotic grasping from object localization pose estimation grasp detection to motion planning: A review. 3"},{"key":"e_1_2_7_12_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-020-09888-5"},{"key":"e_1_2_7_13_2","doi-asserted-by":"crossref","unstructured":"GuptaK. PeterssonL. HartleyR.: Cullnet: Calibrated and pose aware confidence scores for object pose estimation. InProceedings of the IEEE\/CVF International Conference on Computer Vision Workshops(2019) pp.0\u20130. 3","DOI":"10.1109\/ICCVW.2019.00337"},{"key":"e_1_2_7_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2011.206"},{"key":"e_1_2_7_15_2","doi-asserted-by":"crossref","unstructured":"HuY. FuaP. WangW. SalzmannM.: Single-stage 6d object pose estimation.computer vision and pattern recognition(2020). 8","DOI":"10.1109\/CVPR42600.2020.00300"},{"key":"e_1_2_7_16_2","doi-asserted-by":"crossref","unstructured":"HinterstoisserS. HolzerS. J. J. CagniartC. IlicS. KonoligeK. NavabN. LepetitV.: Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes.international conference on computer vision(2011). 6","DOI":"10.1109\/ICCV.2011.6126326"},{"key":"e_1_2_7_17_2","doi-asserted-by":"crossref","unstructured":"HeY. HuangH. FanH. ChenQ. SunJ.: Ffb6d: A full flow bidirectional fusion network for 6d pose estimation. InProceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition(2021) pp.3003\u20133013. 2 3 4 6 7 8 9","DOI":"10.1109\/CVPR46437.2021.00302"},{"key":"e_1_2_7_18_2","doi-asserted-by":"crossref","unstructured":"HuY. HugonotJ. FuaP. SalzmannM.: Segmentation-driven 6d object pose estimation. InProceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition(2019) pp.3385\u20133394. 3","DOI":"10.1109\/CVPR.2019.00350"},{"key":"e_1_2_7_19_2","article-title":"A comprehensive review of 3d point cloud descriptors","volume":"2","author":"Hana X.-F.","year":"2018","journal-title":"arXiv preprint arXiv:1802.02297"},{"key":"e_1_2_7_20_2","first-page":"548","volume-title":"Asian conference on computer vision","author":"Hinterstoisser S.","year":"2012"},{"key":"e_1_2_7_21_2","first-page":"834","volume-title":"European conference on computer vision","author":"Hinterstoisser S.","year":"2016"},{"key":"e_1_2_7_22_2","unstructured":"HuangX. MeiG. ZhangJ. AbbasR.: A comprehensive survey on point cloud registration.arXiv preprint arXiv:2103.02690(2021). 3"},{"key":"e_1_2_7_23_2","doi-asserted-by":"crossref","unstructured":"HeY. SunW. HuangH. LiuJ. FanH. SunJ.: Pvn3d: A deep point-wise 3d keypoints voting network for 6dof pose estimation. InProceedings of the IEEE\/CVF conference on computer vision and pattern recognition(2020) pp.11632\u201311641. 2 3 4 6 8","DOI":"10.1109\/CVPR42600.2020.01165"},{"key":"e_1_2_7_24_2","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2015.7354005"},{"key":"e_1_2_7_25_2","unstructured":"IoffeS. SzegedyC.: Batch normalization: Accelerating deep network training by reducing internal covariate shift.international conference on machine learning(2015). 5"},{"key":"e_1_2_7_26_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2021.02.012"},{"key":"e_1_2_7_27_2","doi-asserted-by":"crossref","unstructured":"KuJ. HarakehA. WaslanderS. L.: In defense of classical image processing: Fast depth completion on the cpu.canadian conference on computer and robot vision(2018). 6","DOI":"10.1109\/CRV.2018.00013"},{"key":"e_1_2_7_28_2","doi-asserted-by":"crossref","unstructured":"KehlW. ManhardtF. TombariF. IlicS. NavabN.: Ssd-6d: Making rgb-based 3d detection and 6d pose estimation great again. InProceedings of the IEEE international conference on computer vision(2017) pp.1521\u20131529. 3","DOI":"10.1109\/ICCV.2017.169"},{"key":"e_1_2_7_29_2","unstructured":"LinT.-Y. GoyalP. GirshickR. HeK. Doll\u00e1rP.: Focal loss for dense object detection.IEEE Transactions on Pattern Analysis and Machine Intelligence(2017). 6"},{"key":"e_1_2_7_30_2","article-title":"Sparse steerable convolutions: An efficient learning of se (3)-equivariant features for estimation and tracking of object poses in 3d space","volume":"34","author":"Lin J.","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_7_31_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-008-0152-6"},{"key":"e_1_2_7_32_2","doi-asserted-by":"crossref","unstructured":"LiY. WangG. JiX. XiangY. FoxD.: Deepim: Deep iterative matching for 6d pose estimation. InProceedings of the European Conference on Computer Vision (ECCV)(2018) pp.683\u2013698. 3 8","DOI":"10.1007\/978-3-030-01231-1_42"},{"key":"e_1_2_7_33_2","article-title":"Leveraging se (3) equivariance for self-supervised category-level object pose estimation from point clouds","volume":"34","author":"Li X.","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_7_34_2","article-title":"A tutorial review on point cloud registrations: Principle, classification, comparison, and technology challenges","volume":"2021","author":"Li L.","year":"2021","journal-title":"Mathematical Problems in Engineering"},{"key":"e_1_2_7_35_2","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2015.2463671"},{"key":"e_1_2_7_36_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-020-01359-2"},{"key":"e_1_2_7_37_2","doi-asserted-by":"crossref","unstructured":"PengS. LiuY. HuangQ. ZhouX. BaoH.: Pvnet: Pixel-wise voting network for 6dof pose estimation. InProceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition(2019) pp.4561\u20134570. 6 8","DOI":"10.1109\/CVPR.2019.00469"},{"key":"e_1_2_7_38_2","doi-asserted-by":"crossref","unstructured":"ParkK. PattenT. VinczeM.: Pix2pose: Pixel-wise coordinate regression of objects for 6d pose estimation.international conference on computer vision(2019). 8","DOI":"10.1109\/ICCV.2019.00776"},{"key":"e_1_2_7_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/3DV.2019.00015"},{"key":"e_1_2_7_40_2","doi-asserted-by":"crossref","unstructured":"PengS. ZhouX. LiuY. LinH. HuangQ. BaoH.: Pvnet: pixel-wise voting network for 6dof object pose estimation.IEEE Transactions on Pattern Analysis and Machine Intelligence(2020). 1 3","DOI":"10.1109\/CVPR.2019.00469"},{"key":"e_1_2_7_41_2","doi-asserted-by":"crossref","unstructured":"Rios-CabreraR. TuytelaarsT.: Discriminatively trained templates for 3d object detection: A real time scalable approach. InProceedings of the IEEE international conference on computer vision(2013) pp.2048\u20132055. 3","DOI":"10.1109\/ICCV.2013.256"},{"key":"e_1_2_7_42_2","doi-asserted-by":"crossref","unstructured":"RadM. LepetitV.: Bb8: A scalable accurate robust to partial occlusion method for predicting the 3d poses of challenging objects without using depth. InProceedings of the IEEE international conference on computer vision(2017) pp.3828\u20133836. 3","DOI":"10.1109\/ICCV.2017.413"},{"key":"e_1_2_7_43_2","doi-asserted-by":"crossref","unstructured":"ShiY. HuangJ. XuX. ZhangY. XuK.: Stablepose: Learning 6d object poses from geometrically stable patches. InProceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition(2021) pp.15222\u201315231. 3","DOI":"10.1109\/CVPR46437.2021.01497"},{"key":"e_1_2_7_44_2","doi-asserted-by":"crossref","unstructured":"SundermeyerM. MartonZ.-C. DurnerM. BruckerM. TriebelR.: Implicit 3d orientation learning for 6d object detection from rgb images. InProceedings of the european conference on computer vision (ECCV)(2018) pp.699\u2013715. 3","DOI":"10.1007\/978-3-030-01231-1_43"},{"key":"e_1_2_7_45_2","doi-asserted-by":"crossref","unstructured":"SongC. SongJ. HuangQ.: Hybridpose: 6d object pose estimation under hybrid representations.computer vision and pattern recognition(2020). 8","DOI":"10.1109\/CVPR42600.2020.00051"},{"key":"e_1_2_7_46_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cviu.2014.04.011"},{"key":"e_1_2_7_47_2","doi-asserted-by":"crossref","unstructured":"TekinB. SinhaS. N. FuaP.: Real-time seamless single shot 6d object pose prediction. InProceedings of the IEEE conference on computer vision and pattern recognition(2018) pp.292\u2013301. 3","DOI":"10.1109\/CVPR.2018.00038"},{"key":"e_1_2_7_48_2","unstructured":"UlyanovD. VedaldiA. LempitskyV.: Instance normalization: The missing ingredient for fast stylization.arXiv: Computer Vision and Pattern Recognition(2016). 5"},{"key":"e_1_2_7_49_2","doi-asserted-by":"publisher","DOI":"10.3390\/s18082678"},{"key":"e_1_2_7_50_2","doi-asserted-by":"crossref","unstructured":"WangC. XuD. ZhuY. Mart\u00edn-Mart\u00ednR. LuC. Fei-FeiL. SavareseS.: Densefusion: 6d object pose estimation by iterative dense fusion.computer vision and pattern recognition(2019). 2 3 6 7 8","DOI":"10.1109\/CVPR.2019.00346"},{"key":"e_1_2_7_51_2","doi-asserted-by":"crossref","unstructured":"WangH. ZhuY. AdamH. YuilleA. ChenL.-C.: Max-deeplab: End-to-end panoptic segmentation with mask transformers. InProceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition(2021) pp.5463\u20135474. 3","DOI":"10.1109\/CVPR46437.2021.00542"},{"key":"e_1_2_7_52_2","doi-asserted-by":"crossref","unstructured":"XuD. AnguelovD. JainA.: Pointfusion: Deep sensor fusion for 3d bounding box estimation.computer vision and pattern recognition(2017). 8","DOI":"10.1109\/CVPR.2018.00033"},{"key":"e_1_2_7_53_2","doi-asserted-by":"crossref","unstructured":"XiangY. SchmidtT. NarayananV. FoxD.: Posecnn: A convolutional neural network for 6d object pose estimation in cluttered scenes.robotics science and systems(2017). 3 6 7 8","DOI":"10.15607\/RSS.2018.XIV.019"},{"key":"e_1_2_7_54_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.robot.2017.06.003"},{"key":"e_1_2_7_55_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cag.2020.05.008"},{"key":"e_1_2_7_56_2","doi-asserted-by":"crossref","unstructured":"ZakharovS. ShugurovI. IlicS.: Dpod: 6d pose object detector and refiner.international conference on computer vision(2019). 8","DOI":"10.1109\/ICCV.2019.00203"},{"key":"e_1_2_7_57_2","doi-asserted-by":"crossref","unstructured":"ZhaoH. ShiJ. QiX. WangX. JiaJ.: Pyramid scene parsing network.computer vision and pattern recognition(2017). 6","DOI":"10.1109\/CVPR.2017.660"}],"container-title":["Computer Graphics Forum"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/cgf.14684","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/cgf.14684","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/cgf.14684","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,8,20]],"date-time":"2023-08-20T22:31:22Z","timestamp":1692570682000},"score":1,"resource":{"primary":{"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/onlinelibrary.wiley.com\/doi\/10.1111\/cgf.14684"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10]]},"references-count":56,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2022,10]]}},"alternative-id":["10.1111\/cgf.14684"],"URL":"https:\/\/2.zoppoz.workers.dev:443\/https\/doi.org\/10.1111\/cgf.14684","archive":["Portico"],"relation":{},"ISSN":["0167-7055","1467-8659"],"issn-type":[{"value":"0167-7055","type":"print"},{"value":"1467-8659","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,10]]},"assertion":[{"value":"2023-03-20","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}