{"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/mono-camera-3d-multi-object-tracking-using","title":"Mono-Camera 3D Multi-Object Tracking Using Deep Learning Detections and PMBM Filtering","arxiv_id":"1802.09975","date":"2018-02-27","proceeding":null,"authors":["Samuel Scheidegger","Joachim Benjaminsson","Emil Rosenberg","Amrit Krishnan","Karl Granstrom"],"abstract":"Monocular cameras are one of the most commonly used sensors in the automotive\nindustry for autonomous vehicles. One major drawback using a monocular camera\nis that it only makes observations in the two dimensional image plane and can\nnot directly measure the distance to objects. In this paper, we aim at filling\nthis gap by developing a multi-object tracking algorithm that takes an image as\ninput and produces trajectories of detected objects in a world coordinate\nsystem. We solve this by using a deep neural network trained to detect and\nestimate the distance to objects from a single input image. The detections from\na sequence of images are fed in to a state-of-the art Poisson multi-Bernoulli\nmixture tracking filter. The combination of the learned detector and the PMBM\nfilter results in an algorithm that achieves 3D tracking using only mono-camera\nimages as input. The performance of the algorithm is evaluated both in 3D world\ncoordinates, and 2D image coordinates, using the publicly available KITTI\nobject tracking dataset. The algorithm shows the ability to accurately track\nobjects, correctly handle data associations, even when there is a big overlap\nof the objects in the image, and is one of the top performing algorithms on the\nKITTI object tracking benchmark. Furthermore, the algorithm is efficient,\nrunning on average close to 20 frames per second.","url_abs":"http://arxiv.org/abs/1802.09975v1","url_pdf":"http://arxiv.org/pdf/1802.09975v1.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":"3d-multi-object-tracking","task_name":"3D Multi-Object Tracking"},{"task_slug":"autonomous-vehicles","task_name":"Autonomous Vehicles"},{"task_slug":"multi-object-tracking","task_name":"Multi-Object Tracking"},{"task_slug":"multiple-object-tracking","task_name":"Multiple Object Tracking"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-tracking","task_name":"Object Tracking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/multiple-object-tracking-on-kitti-test-online","task":"Multiple Object Tracking","dataset":"KITTI Test (Online Methods)","model":"PMBM","rank_in_archive_order":26,"of":34,"metrics":{"MOTA":"80.39"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.09975","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}