Corentin Sautier, Gilles Puy, Alexandre Boulch, Renaud Marlet, Vincent Lepetit
{"title":"单元:通过时间进行无监督在线实例分割","authors":"Corentin Sautier, Gilles Puy, Alexandre Boulch, Renaud Marlet, Vincent Lepetit","doi":"arxiv-2409.07887","DOIUrl":null,"url":null,"abstract":"Online object segmentation and tracking in Lidar point clouds enables\nautonomous agents to understand their surroundings and make safe decisions.\nUnfortunately, manual annotations for these tasks are prohibitively costly. We\ntackle this problem with the task of class-agnostic unsupervised online\ninstance segmentation and tracking. To that end, we leverage an instance\nsegmentation backbone and propose a new training recipe that enables the online\ntracking of objects. Our network is trained on pseudo-labels, eliminating the\nneed for manual annotations. We conduct an evaluation using metrics adapted for\ntemporal instance segmentation. Computing these metrics requires\ntemporally-consistent instance labels. When unavailable, we construct these\nlabels using the available 3D bounding boxes and semantic labels in the\ndataset. We compare our method against strong baselines and demonstrate its\nsuperiority across two different outdoor Lidar datasets.","PeriodicalId":501130,"journal":{"name":"arXiv - CS - Computer Vision and Pattern Recognition","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"UNIT: Unsupervised Online Instance Segmentation through Time\",\"authors\":\"Corentin Sautier, Gilles Puy, Alexandre Boulch, Renaud Marlet, Vincent Lepetit\",\"doi\":\"arxiv-2409.07887\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Online object segmentation and tracking in Lidar point clouds enables\\nautonomous agents to understand their surroundings and make safe decisions.\\nUnfortunately, manual annotations for these tasks are prohibitively costly. We\\ntackle this problem with the task of class-agnostic unsupervised online\\ninstance segmentation and tracking. To that end, we leverage an instance\\nsegmentation backbone and propose a new training recipe that enables the online\\ntracking of objects. Our network is trained on pseudo-labels, eliminating the\\nneed for manual annotations. We conduct an evaluation using metrics adapted for\\ntemporal instance segmentation. Computing these metrics requires\\ntemporally-consistent instance labels. When unavailable, we construct these\\nlabels using the available 3D bounding boxes and semantic labels in the\\ndataset. We compare our method against strong baselines and demonstrate its\\nsuperiority across two different outdoor Lidar datasets.\",\"PeriodicalId\":501130,\"journal\":{\"name\":\"arXiv - CS - Computer Vision and Pattern Recognition\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Computer Vision and Pattern Recognition\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.07887\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Computer Vision and Pattern Recognition","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07887","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
UNIT: Unsupervised Online Instance Segmentation through Time
Online object segmentation and tracking in Lidar point clouds enables
autonomous agents to understand their surroundings and make safe decisions.
Unfortunately, manual annotations for these tasks are prohibitively costly. We
tackle this problem with the task of class-agnostic unsupervised online
instance segmentation and tracking. To that end, we leverage an instance
segmentation backbone and propose a new training recipe that enables the online
tracking of objects. Our network is trained on pseudo-labels, eliminating the
need for manual annotations. We conduct an evaluation using metrics adapted for
temporal instance segmentation. Computing these metrics requires
temporally-consistent instance labels. When unavailable, we construct these
labels using the available 3D bounding boxes and semantic labels in the
dataset. We compare our method against strong baselines and demonstrate its
superiority across two different outdoor Lidar datasets.