{"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/neural-best-buddies-sparse-cross-domain","title":"Neural Best-Buddies: Sparse Cross-Domain Correspondence","arxiv_id":"1805.04140","date":"2018-05-10","proceeding":null,"authors":["Kfir Aberman","Jing Liao","Mingyi Shi","Dani Lischinski","Baoquan Chen","Daniel Cohen-Or"],"abstract":"Correspondence between images is a fundamental problem in computer vision,\nwith a variety of graphics applications. This paper presents a novel method for\nsparse cross-domain correspondence. Our method is designed for pairs of images\nwhere the main objects of interest may belong to different semantic categories\nand differ drastically in shape and appearance, yet still contain semantically\nrelated or geometrically similar parts. Our approach operates on hierarchies of\ndeep features, extracted from the input images by a pre-trained CNN.\nSpecifically, starting from the coarsest layer in both hierarchies, we search\nfor Neural Best Buddies (NBB): pairs of neurons that are mutual nearest\nneighbors. The key idea is then to percolate NBBs through the hierarchy, while\nnarrowing down the search regions at each level and retaining only NBBs with\nsignificant activations. Furthermore, in order to overcome differences in\nappearance, each pair of search regions is transformed into a common\nappearance. We evaluate our method via a user study, in addition to comparisons\nwith alternative correspondence approaches. The usefulness of our method is\ndemonstrated using a variety of graphics applications, including cross-domain\nimage alignment, creation of hybrid images, automatic image morphing, and more.","url_abs":"http://arxiv.org/abs/1805.04140v2","url_pdf":"http://arxiv.org/pdf/1805.04140v2.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":"neural-best-buddies-sparse-cross-domain","repo_url":"https://github.com/kfiraberman/neural_best_buddies","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"neural-best-buddies-sparse-cross-domain","repo_url":"https://github.com/sunniesuhyoung/DST","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-morphing","task_name":"Image Morphing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}