{"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/ps-fcn-a-flexible-learning-framework-for","title":"PS-FCN: A Flexible Learning Framework for Photometric Stereo","arxiv_id":"1807.08696","date":"2018-07-23","proceeding":"ECCV 2018 9","authors":["Guan-Ying Chen","Kai Han","Kwan-Yee K. Wong"],"abstract":"This paper addresses the problem of photometric stereo for non-Lambertian\nsurfaces. Existing approaches often adopt simplified reflectance models to make\nthe problem more tractable, but this greatly hinders their applications on\nreal-world objects. In this paper, we propose a deep fully convolutional\nnetwork, called PS-FCN, that takes an arbitrary number of images of a static\nobject captured under different light directions with a fixed camera as input,\nand predicts a normal map of the object in a fast feed-forward pass. Unlike the\nrecently proposed learning based method, PS-FCN does not require a pre-defined\nset of light directions during training and testing, and can handle multiple\nimages and light directions in an order-agnostic manner. Although we train\nPS-FCN on synthetic data, it can generalize well on real datasets. We further\nshow that PS-FCN can be easily extended to handle the problem of uncalibrated\nphotometric stereo.Extensive experiments on public real datasets show that\nPS-FCN outperforms existing approaches in calibrated photometric stereo, and\npromising results are achieved in uncalibrated scenario, clearly demonstrating\nits effectiveness.","url_abs":"http://arxiv.org/abs/1807.08696v1","url_pdf":"http://arxiv.org/pdf/1807.08696v1.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":"ps-fcn-a-flexible-learning-framework-for","repo_url":"https://github.com/guanyingc/PS-FCN","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.08696","atlas_url":"https://app.syntology.ai/?focus=1807.08696","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}