{"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/virtual-worlds-as-proxy-for-multi-object","title":"Virtual Worlds as Proxy for Multi-Object Tracking Analysis","arxiv_id":"1605.06457","date":"2016-05-20","proceeding":"CVPR 2016 6","authors":["Adrien Gaidon","Qiao Wang","Yohann Cabon","Eleonora Vig"],"abstract":"Modern computer vision algorithms typically require expensive data\nacquisition and accurate manual labeling. In this work, we instead leverage the\nrecent progress in computer graphics to generate fully labeled, dynamic, and\nphoto-realistic proxy virtual worlds. We propose an efficient real-to-virtual\nworld cloning method, and validate our approach by building and publicly\nreleasing a new video dataset, called Virtual KITTI (see\nhttp://www.xrce.xerox.com/Research-Development/Computer-Vision/Proxy-Virtual-Worlds),\nautomatically labeled with accurate ground truth for object detection,\ntracking, scene and instance segmentation, depth, and optical flow. We provide\nquantitative experimental evidence suggesting that (i) modern deep learning\nalgorithms pre-trained on real data behave similarly in real and virtual\nworlds, and (ii) pre-training on virtual data improves performance. As the gap\nbetween real and virtual worlds is small, virtual worlds enable measuring the\nimpact of various weather and imaging conditions on recognition performance,\nall other things being equal. We show these factors may affect drastically\notherwise high-performing deep models for tracking.","url_abs":"http://arxiv.org/abs/1605.06457v1","url_pdf":"http://arxiv.org/pdf/1605.06457v1.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":[],"tasks":[{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"multi-object-tracking","task_name":"Multi-Object Tracking"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[{"slug":"virtual-kitti","name":"Virtual KITTI","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1605.06457","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}