{"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/reverse-attention-for-salient-object","title":"Reverse Attention for Salient Object Detection","arxiv_id":"1807.09940","date":"2018-07-26","proceeding":"ECCV 2018 9","authors":["Shuhan Chen","Xiuli Tan","Ben Wang","Xuelong Hu"],"abstract":"Benefit from the quick development of deep learning techniques, salient\nobject detection has achieved remarkable progresses recently. However, there\nstill exists following two major challenges that hinder its application in\nembedded devices, low resolution output and heavy model weight. To this end,\nthis paper presents an accurate yet compact deep network for efficient salient\nobject detection. More specifically, given a coarse saliency prediction in the\ndeepest layer, we first employ residual learning to learn side-output residual\nfeatures for saliency refinement, which can be achieved with very limited\nconvolutional parameters while keep accuracy. Secondly, we further propose\nreverse attention to guide such side-output residual learning in a top-down\nmanner. By erasing the current predicted salient regions from side-output\nfeatures, the network can eventually explore the missing object parts and\ndetails which results in high resolution and accuracy. Experiments on six\nbenchmark datasets demonstrate that the proposed approach compares favorably\nagainst state-of-the-art methods, and with advantages in terms of simplicity,\nefficiency (45 FPS) and model size (81 MB).","url_abs":"http://arxiv.org/abs/1807.09940v2","url_pdf":"http://arxiv.org/pdf/1807.09940v2.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":"reverse-attention-for-salient-object","repo_url":"https://github.com/ShuhanChen/RAS_ECCV18","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"caffe2","reach":null},{"paper_slug":"reverse-attention-for-salient-object","repo_url":"https://github.com/lhaof/Salient-Object-Detection-Papers","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"reverse-attention-for-salient-object","repo_url":"https://github.com/lhaof/fast-salient-object-detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"salient-object-detection","task_name":"RGB Salient Object Detection"},{"task_slug":"saliency-prediction","task_name":"Saliency Prediction"},{"task_slug":"salient-object-detection-1","task_name":"Salient Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.09940","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}