{"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/a-framework-for-evaluating-6-dof-object","title":"A Framework for Evaluating 6-DOF Object Trackers","arxiv_id":"1803.10075","date":"2018-03-27","proceeding":"ECCV 2018 9","authors":["Mathieu Garon","Denis Laurendeau","Jean-François Lalonde"],"abstract":"We present a challenging and realistic novel dataset for evaluating 6-DOF\nobject tracking algorithms. Existing datasets show serious\nlimitations---notably, unrealistic synthetic data, or real data with large\nfiducial markers---preventing the community from obtaining an accurate picture\nof the state-of-the-art. Using a data acquisition pipeline based on a\ncommercial motion capture system for acquiring accurate ground truth poses of\nreal objects with respect to a Kinect V2 camera, we build a dataset which\ncontains a total of 297 calibrated sequences. They are acquired in three\ndifferent scenarios to evaluate the performance of trackers: stability,\nrobustness to occlusion and accuracy during challenging interactions between a\nperson and the object. We conduct an extensive study of a deep 6-DOF tracking\narchitecture and determine a set of optimal parameters. We enhance the\narchitecture and the training methodology to train a 6-DOF tracker that can\nrobustly generalize to objects never seen during training, and demonstrate\nfavorable performance compared to previous approaches trained specifically on\nthe objects to track.","url_abs":"http://arxiv.org/abs/1803.10075v3","url_pdf":"http://arxiv.org/pdf/1803.10075v3.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":"a-framework-for-evaluating-6-dof-object","repo_url":"https://github.com/lvsn/6DOF_tracking_evaluation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-tracking","task_name":"Object Tracking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1803.10075","atlas_url":"https://app.syntology.ai/?focus=1803.10075","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}