{"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/learning-uncertain-convolutional-features-for","title":"Learning Uncertain Convolutional Features for Accurate Saliency Detection","arxiv_id":"1708.02031","date":"2017-08-07","proceeding":"ICCV 2017 10","authors":["Pingping Zhang","Dong Wang","Huchuan Lu","Hongyu Wang","Bao-Cai Yin"],"abstract":"Deep convolutional neural networks (CNNs) have delivered superior performance\nin many computer vision tasks. In this paper, we propose a novel deep fully\nconvolutional network model for accurate salient object detection. The key\ncontribution of this work is to learn deep uncertain convolutional features\n(UCF), which encourage the robustness and accuracy of saliency detection. We\nachieve this via introducing a reformulated dropout (R-dropout) after specific\nconvolutional layers to construct an uncertain ensemble of internal feature\nunits. In addition, we propose an effective hybrid upsampling method to reduce\nthe checkerboard artifacts of deconvolution operators in our decoder network.\nThe proposed methods can also be applied to other deep convolutional networks.\nCompared with existing saliency detection methods, the proposed UCF model is\nable to incorporate uncertainties for more accurate object boundary inference.\nExtensive experiments demonstrate that our proposed saliency model performs\nfavorably against state-of-the-art approaches. The uncertain feature learning\nmechanism as well as the upsampling method can significantly improve\nperformance on other pixel-wise vision tasks.","url_abs":"http://arxiv.org/abs/1708.02031v1","url_pdf":"http://arxiv.org/pdf/1708.02031v1.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":"learning-uncertain-convolutional-features-for","repo_url":"https://github.com/Pchank/caffe-sal","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"salient-object-detection","task_name":"RGB Salient Object Detection"},{"task_slug":"saliency-detection","task_name":"Saliency Detection"},{"task_slug":"salient-object-detection-1","task_name":"Salient Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"dropout","method_name":"Dropout"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/salient-object-detection-on-duts-te","task":"RGB Salient Object Detection","dataset":"DUTS-TE","model":"UCF","rank_in_archive_order":28,"of":31,"metrics":{"MAE":"0.116","max F-measure":"0.771"},"uses_additional_data":false},{"leaderboard":"/sota/saliency-detection-on-dut-omron","task":"Saliency Detection","dataset":"DUT-OMRON","model":"UCF","rank_in_archive_order":5,"of":5,"metrics":{"MAE":"0.1203"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1708.02031","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}