{"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/joint-learning-of-intrinsic-images-and","title":"Joint Learning of Intrinsic Images and Semantic Segmentation","arxiv_id":"1807.11857","date":"2018-07-31","proceeding":"ECCV 2018 9","authors":["Anil S. Baslamisli","Thomas T. Groenestege","Partha Das","Hoang-An Le","Sezer Karaoglu","Theo Gevers"],"abstract":"Semantic segmentation of outdoor scenes is problematic when there are\nvariations in imaging conditions. It is known that albedo (reflectance) is\ninvariant to all kinds of illumination effects. Thus, using reflectance images\nfor semantic segmentation task can be favorable. Additionally, not only\nsegmentation may benefit from reflectance, but also segmentation may be useful\nfor reflectance computation. Therefore, in this paper, the tasks of semantic\nsegmentation and intrinsic image decomposition are considered as a combined\nprocess by exploring their mutual relationship in a joint fashion. To that end,\nwe propose a supervised end-to-end CNN architecture to jointly learn intrinsic\nimage decomposition and semantic segmentation. We analyze the gains of\naddressing those two problems jointly. Moreover, new cascade CNN architectures\nfor intrinsic-for-segmentation and segmentation-for-intrinsic are proposed as\nsingle tasks. Furthermore, a dataset of 35K synthetic images of natural\nenvironments is created with corresponding albedo and shading (intrinsics), as\nwell as semantic labels (segmentation) assigned to each object/scene. The\nexperiments show that joint learning of intrinsic image decomposition and\nsemantic segmentation is beneficial for both tasks for natural scenes. Dataset\nand models are available at: https://ivi.fnwi.uva.nl/cv/intrinseg","url_abs":"http://arxiv.org/abs/1807.11857v1","url_pdf":"http://arxiv.org/pdf/1807.11857v1.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":"joint-learning-of-intrinsic-images-and","repo_url":"https://github.com/Morpheus3000/intrinseg","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"intrinsic-image-decomposition","task_name":"Intrinsic Image Decomposition"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.11857","atlas_url":"https://app.syntology.ai/?focus=1807.11857","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}