{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T10:45:03Z","timestamp":1775040303270,"version":"3.50.1"},"reference-count":69,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,5,29]],"date-time":"2023-05-29T00:00:00Z","timestamp":1685318400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,5,29]],"date-time":"2023-05-29T00:00:00Z","timestamp":1685318400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,5,29]]},"DOI":"10.1109\/icra48891.2023.10160846","type":"proceedings-article","created":{"date-parts":[[2023,7,4]],"date-time":"2023-07-04T17:20:56Z","timestamp":1688491256000},"page":"9132-9139","source":"Crossref","is-referenced-by-count":30,"title":["ARMBench: An Object-centric Benchmark Dataset for Robotic Manipulation"],"prefix":"10.1109","author":[{"given":"Chaitanya","family":"Mitash","sequence":"first","affiliation":[{"name":"Amazon Robotics,MA,USA"}]},{"given":"Fan","family":"Wang","sequence":"additional","affiliation":[{"name":"Amazon Robotics,MA,USA"}]},{"given":"Shiyang","family":"Lu","sequence":"additional","affiliation":[{"name":"Rutgers University,Computer Science Department,NJ,USA"}]},{"given":"Vikedo","family":"Terhuja","sequence":"additional","affiliation":[{"name":"Amazon Robotics,MA,USA"}]},{"given":"Tyler","family":"Garaas","sequence":"additional","affiliation":[{"name":"Amazon Robotics,MA,USA"}]},{"given":"Felipe","family":"Polido","sequence":"additional","affiliation":[{"name":"Amazon Robotics,MA,USA"}]},{"given":"Manikantan","family":"Nambi","sequence":"additional","affiliation":[{"name":"Amazon Robotics,MA,USA"}]}],"member":"263","reference":[{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58577-8_19"},{"key":"ref57","article-title":"Object tracking by detection with visual and motion cues","volume":"abs 2101 7549","author":"salscheider","year":"2021","journal-title":"CoRR"},{"key":"ref12","article-title":"Shapenet: An information-rich 3d model repository","author":"chang","year":"2015","journal-title":"ArXiv Preprint"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.02045"},{"key":"ref59","first-page":"1470","article-title":"Video google: A text retrieval approach to object matching in videos","author":"sivic","year":"2003","journal-title":"null IEEE"},{"key":"ref14","author":"colling","year":"2018","journal-title":"Progress in autonomous picking as demonstrated by the amazon robotic challenge"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.2478\/aut-2019-0035"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2846566"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2008.4587635"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00951"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2016.2532924"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1177\/0278364917700714"},{"key":"ref54","article-title":"End-to-end instance segmentation and counting with recurrent attention","volume":"abs 1605 9410","author":"ren","year":"2016","journal-title":"CoRR"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/TASE.2016.2600527"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.350"},{"key":"ref19","article-title":"Segmenting unknown 3d objects from real depth images using mask R-CNN trained on synthetic point clouds","volume":"abs 1809 5825","author":"danielczuk","year":"2018","journal-title":"CoRR"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/TASE.2016.2600527"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2007.383172"},{"key":"ref50","article-title":"Rp2k: A large-scale retail product dataset for fine-grained image classification","author":"peng","year":"2020","journal-title":"ArXiv Preprint"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00684"},{"key":"ref45","article-title":"Microsoft coco: Common objects in context","author":"lin","year":"2014","journal-title":"ECCV"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2017.XIII.058"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.124"},{"key":"ref42","first-page":"18","article-title":"Anomaly detection and localization in crowded scenes","volume":"36","author":"li","year":"2013","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2017.7989545"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1007\/s10845-018-1415-x"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2010.2040278"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.434"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10605-2_35"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00982"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2021.3114936"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10590-1_38"},{"key":"ref3","author":"abdulla","year":"2017","journal-title":"Mask r-cnn for object detection and instance segmentation on keras and tensorflow"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TIM.2021.3083561"},{"key":"ref5","article-title":"Products-10 k: A large-scale product recognition dataset","author":"bai","year":"2020","journal-title":"ArXiv Preprint"},{"key":"ref40","article-title":"Anomaly locality in video surveillance","author":"landi","year":"2019","journal-title":"ArXiv Preprint"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/WACV.2017.103"},{"key":"ref34","article-title":"Deep residual learning for image recognition","volume":"abs 1512 3385","author":"he","year":"2015","journal-title":"CoRR"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2010.5540039"},{"key":"ref36","first-page":"19","article-title":"Bop: Benchmark for 6d object pose estimation","author":"hodan","year":"2018","journal-title":"Proceedings of the European Conference on Computer Vision (ECCV)"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00548"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1016\/j.imavis.2019.01.001"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref32","article-title":"Bicameral structuring and synthetic imagery for jointly predicting instance boundaries and nearby occlusions from a single image","volume":"abs 1906 7480","author":"grard","year":"2019","journal-title":"CoRR"},{"key":"ref2","year":"2022","journal-title":"DAGM 2007"},{"key":"ref1","year":"0","journal-title":"Amazon introduces sparrow—a state-of-the-art robot that handles millions of diverse products"},{"key":"ref39","article-title":"The kinetics human action video dataset","author":"kay","year":"2017","journal-title":"ArXiv Preprint"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.223"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2017\/676"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2017.7989165"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA46639.2022.9811809"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2021.3060341"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00675"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-009-0275-4"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.79"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2019.8793744"},{"key":"ref64","article-title":"Rpc: A large-scale retail product checkout dataset","author":"wei","year":"2019","journal-title":"ArXiv Preprint"},{"key":"ref63","article-title":"Deep object pose estimation for semantic robotic grasping of household objects","volume":"abs 1809 10790","author":"tremblay","year":"2018","journal-title":"CoRR"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58558-7_35"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2018.XIV.019"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW.2017.66"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00265"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01249-6_35"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/7287.001.0001"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2017.330"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0810-4"},{"key":"ref62","article-title":"Particular object retrieval with integral max-pooling of cnn activations","author":"tolias","year":"2015","journal-title":"ArXiv Preprint"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_27"}],"event":{"name":"2023 IEEE International Conference on Robotics and Automation (ICRA)","location":"London, United Kingdom","start":{"date-parts":[[2023,5,29]]},"end":{"date-parts":[[2023,6,2]]}},"container-title":["2023 IEEE International Conference on Robotics and Automation (ICRA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10160211\/10160212\/10160846.pdf?arnumber=10160846","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,7,24]],"date-time":"2023-07-24T17:33:49Z","timestamp":1690220029000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10160846\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,5,29]]},"references-count":69,"URL":"https:\/\/doi.org\/10.1109\/icra48891.2023.10160846","relation":{},"subject":[],"published":{"date-parts":[[2023,5,29]]}}}