{"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-single-view-3d-object-reconstruction","title":"Deep Single-View 3D Object Reconstruction with Visual Hull Embedding","arxiv_id":"1809.03451","date":"2018-09-10","proceeding":null,"authors":["Hanqing Wang","Jiaolong Yang","Wei Liang","Xin Tong"],"abstract":"3D object reconstruction is a fundamental task of many robotics and AI\nproblems. With the aid of deep convolutional neural networks (CNNs), 3D object\nreconstruction has witnessed a significant progress in recent years. However,\npossibly due to the prohibitively high dimension of the 3D object space, the\nresults from deep CNNs are often prone to missing some shape details. In this\npaper, we present an approach which aims to preserve more shape details and\nimprove the reconstruction quality. The key idea of our method is to leverage\nobject mask and pose estimation from CNNs to assist the 3D shape learning by\nconstructing a probabilistic single-view visual hull inside of the network. Our\nmethod works by first predicting a coarse shape as well as the object pose and\nsilhouette using CNNs, followed by a novel 3D refinement CNN which refines the\ncoarse shapes using the constructed probabilistic visual hulls. Experiment on\nboth synthetic data and real images show that embedding a single-view visual\nhull for shape refinement can significantly improve the reconstruction quality\nby recovering more shapes details and improving shape consistency with the\ninput image.","url_abs":"http://arxiv.org/abs/1809.03451v1","url_pdf":"http://arxiv.org/pdf/1809.03451v1.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":"deep-single-view-3d-object-reconstruction","repo_url":"https://github.com/qweas120/PSVH-3d-reconstruction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"3d-object-reconstruction","task_name":"3D Object Reconstruction"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-reconstruction","task_name":"Object Reconstruction"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}