{"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/generalized-deep-image-to-image-regression","title":"Generalized Deep Image to Image Regression","arxiv_id":"1612.03268","date":"2016-12-10","proceeding":"CVPR 2017 7","authors":["Venkataraman Santhanam","Vlad I. Morariu","Larry S. Davis"],"abstract":"We present a Deep Convolutional Neural Network architecture which serves as a\ngeneric image-to-image regressor that can be trained end-to-end without any\nfurther machinery. Our proposed architecture: the Recursively Branched\nDeconvolutional Network (RBDN) develops a cheap multi-context image\nrepresentation very early on using an efficient recursive branching scheme with\nextensive parameter sharing and learnable upsampling. This multi-context\nrepresentation is subjected to a highly non-linear locality preserving\ntransformation by the remainder of our network comprising of a series of\nconvolutions/deconvolutions without any spatial downsampling. The RBDN\narchitecture is fully convolutional and can handle variable sized images during\ninference. We provide qualitative/quantitative results on $3$ diverse tasks:\nrelighting, denoising and colorization and show that our proposed RBDN\narchitecture obtains comparable results to the state-of-the-art on each of\nthese tasks when used off-the-shelf without any post processing or\ntask-specific architectural modifications.","url_abs":"http://arxiv.org/abs/1612.03268v1","url_pdf":"http://arxiv.org/pdf/1612.03268v1.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":"generalized-deep-image-to-image-regression","repo_url":"https://github.com/venkai/RBDN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"colorization","task_name":"Colorization"},{"task_slug":"denoising","task_name":"Denoising"},{"task_slug":"image-to-image-regression","task_name":"Image-to-Image Regression"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[{"method_slug":"colorization","method_name":"Colorization"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}