{"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/reflectance-adaptive-filtering-improves","title":"Reflectance Adaptive Filtering Improves Intrinsic Image Estimation","arxiv_id":"1612.05062","date":"2016-12-15","proceeding":"CVPR 2017 7","authors":["Thomas Nestmeyer","Peter V. Gehler"],"abstract":"Separating an image into reflectance and shading layers poses a challenge for\nlearning approaches because no large corpus of precise and realistic ground\ntruth decompositions exists. The Intrinsic Images in the Wild~(IIW) dataset\nprovides a sparse set of relative human reflectance judgments, which serves as\na standard benchmark for intrinsic images. A number of methods use IIW to learn\nstatistical dependencies between the images and their reflectance layer.\nAlthough learning plays an important role for high performance, we show that a\nstandard signal processing technique achieves performance on par with current\nstate-of-the-art. We propose a loss function for CNN learning of dense\nreflectance predictions. Our results show a simple pixel-wise decision, without\nany context or prior knowledge, is sufficient to provide a strong baseline on\nIIW. This sets a competitive baseline which only two other approaches surpass.\nWe then develop a joint bilateral filtering method that implements strong prior\nknowledge about reflectance constancy. This filtering operation can be applied\nto any intrinsic image algorithm and we improve several previous results\nachieving a new state-of-the-art on IIW. Our findings suggest that the effect\nof learning-based approaches may have been over-estimated so far. Explicit\nprior knowledge is still at least as important to obtain high performance in\nintrinsic image decompositions.","url_abs":"http://arxiv.org/abs/1612.05062v2","url_pdf":"http://arxiv.org/pdf/1612.05062v2.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":"reflectance-adaptive-filtering-improves","repo_url":"https://github.com/tnestmeyer/reflectance-filtering","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"caffe2","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.05062","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}