{"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/deep-manta-a-coarse-to-fine-many-task-network","title":"Deep MANTA: A Coarse-to-fine Many-Task Network for joint 2D and 3D vehicle analysis from monocular image","arxiv_id":"1703.07570","date":"2017-03-22","proceeding":"CVPR 2017 7","authors":["Florian Chabot","Mohamed Chaouch","Jaonary Rabarisoa","Céline Teulière","Thierry Chateau"],"abstract":"In this paper, we present a novel approach, called Deep MANTA (Deep\nMany-Tasks), for many-task vehicle analysis from a given image. A robust\nconvolutional network is introduced for simultaneous vehicle detection, part\nlocalization, visibility characterization and 3D dimension estimation. Its\narchitecture is based on a new coarse-to-fine object proposal that boosts the\nvehicle detection. Moreover, the Deep MANTA network is able to localize vehicle\nparts even if these parts are not visible. In the inference, the network's\noutputs are used by a real time robust pose estimation algorithm for fine\norientation estimation and 3D vehicle localization. We show in experiments that\nour method outperforms monocular state-of-the-art approaches on vehicle\ndetection, orientation and 3D location tasks on the very challenging KITTI\nbenchmark.","url_abs":"http://arxiv.org/abs/1703.07570v1","url_pdf":"http://arxiv.org/pdf/1703.07570v1.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":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"vehicle-pose-estimation","task_name":"Vehicle Pose Estimation"},{"task_slug":"vehicle-detection","task_name":"vehicle detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/vehicle-pose-estimation-on-kitti-cars-hard","task":"Vehicle Pose Estimation","dataset":"KITTI Cars Hard","model":"Deep-Manta","rank_in_archive_order":2,"of":19,"metrics":{"Average Orientation Similarity":"80.39"},"uses_additional_data":true}],"syntology":{"syntology_url":"https://syntology.ai/paper/1703.07570","atlas_url":"https://app.syntology.ai/?focus=1703.07570","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}