{"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/the-secrets-of-salient-object-segmentation","title":"The Secrets of Salient Object Segmentation","arxiv_id":"1406.2807","date":"2014-06-11","proceeding":"CVPR 2014 6","authors":["Yin Li","Xiaodi Hou","Christof Koch","James M. Rehg","Alan L. Yuille"],"abstract":"In this paper we provide an extensive evaluation of fixation prediction and\nsalient object segmentation algorithms as well as statistics of major datasets.\nOur analysis identifies serious design flaws of existing salient object\nbenchmarks, called the dataset design bias, by over emphasizing the\nstereotypical concepts of saliency. The dataset design bias does not only\ncreate the discomforting disconnection between fixations and salient object\nsegmentation, but also misleads the algorithm designing. Based on our analysis,\nwe propose a new high quality dataset that offers both fixation and salient\nobject segmentation ground-truth. With fixations and salient object being\npresented simultaneously, we are able to bridge the gap between fixations and\nsalient objects, and propose a novel method for salient object segmentation.\nFinally, we report significant benchmark progress on three existing datasets of\nsegmenting salient objects","url_abs":"http://arxiv.org/abs/1406.2807v2","url_pdf":"http://arxiv.org/pdf/1406.2807v2.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":"the-secrets-of-salient-object-segmentation","repo_url":"https://github.com/gqding/salfbnet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[],"datasets_introduced":[{"slug":"pascal-s","name":"PASCAL-S","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1406.2807","atlas_url":"https://app.syntology.ai/?focus=1406.2807","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}