{"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/multichannel-semantic-segmentation-with","title":"Multichannel Semantic Segmentation with Unsupervised Domain Adaptation","arxiv_id":"1812.04351","date":"2018-12-11","proceeding":null,"authors":["Kohei Watanabe","Kuniaki Saito","Yoshitaka Ushiku","Tatsuya Harada"],"abstract":"Most contemporary robots have depth sensors, and research on semantic\nsegmentation with RGBD images has shown that depth images boost the accuracy of\nsegmentation. Since it is time-consuming to annotate images with semantic\nlabels per pixel, it would be ideal if we could avoid this laborious work by\nutilizing an existing dataset or a synthetic dataset which we can generate on\nour own. Robot motions are often tested in a synthetic environment, where\nmultichannel (eg, RGB + depth + instance boundary) images plus their\npixel-level semantic labels are available. However, models trained simply on\nsynthetic images tend to demonstrate poor performance on real images. In order\nto address this, we propose two approaches that can efficiently exploit\nmultichannel inputs combined with an unsupervised domain adaptation (UDA)\nalgorithm. One is a fusion-based approach that uses depth images as inputs. The\nother is a multitask learning approach that uses depth images as outputs. We\ndemonstrated that the segmentation results were improved by using a multitask\nlearning approach with a post-process and created a benchmark for this task.","url_abs":"http://arxiv.org/abs/1812.04351v1","url_pdf":"http://arxiv.org/pdf/1812.04351v1.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":"multichannel-semantic-segmentation-with","repo_url":"https://github.com/LittleWat/multichannel-semseg-with-uda","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"unsupervised-domain-adaptation","task_name":"Unsupervised Domain Adaptation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}