FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2954781872> ?p ?o ?g. }

• W2954781872 abstract "Albeit intensively studied, false prediction and unclear boundaries are still major issues of salient object detection. In this paper, we propose a Region Refinement Network (RRN), which recurrently filters redundant information and explicitly models boundary information for saliency detection. Different from existing refinement methods, we propose a Region Refinement Module (RRM) that optimizes salient region prediction by incorporating supervised attention masks in the intermediate refinement stages. The module only brings a minor increase in model size and yet significantly reduces false predictions from the background. To further refine boundary areas, we propose a Boundary Refinement Loss (BRL) that adds extra supervision for better distinguishing foreground from background. BRL is parameter free and easy to train. We further observe that BRL helps retain the integrity in prediction by refining the boundary. Extensive experiments on saliency detection datasets show that our refinement module and loss bring significant improvement to the baseline and can be easily applied to different frameworks. We also demonstrate that our proposed model generalizes well to portrait segmentation and shadow detection tasks." @default.
• W2954781872 created "2019-07-12" @default.
• W2954781872 creator A5038086218 @default.
• W2954781872 creator A5042926447 @default.
• W2954781872 creator A5052856441 @default.
• W2954781872 creator A5061180684 @default.
• W2954781872 creator A5063642139 @default.
• W2954781872 creator A5078109015 @default.
• W2954781872 creator A5079917583 @default.
• W2954781872 date "2019-06-27" @default.
• W2954781872 modified "2023-09-23" @default.
• W2954781872 title "Region Refinement Network for Salient Object Detection" @default.
• W2954781872 cites W1894057436 @default.
• W2954781872 cites W1897243830 @default.
• W2954781872 cites W1942214758 @default.
• W2954781872 cites W1947031653 @default.
• W2954781872 cites W1996326832 @default.
• W2954781872 cites W2002781701 @default.
• W2954781872 cites W2027850463 @default.
• W2954781872 cites W2037954058 @default.
• W2954781872 cites W2039313011 @default.
• W2954781872 cites W2047670868 @default.
• W2954781872 cites W2059753722 @default.
• W2954781872 cites W2086791339 @default.
• W2954781872 cites W2100470808 @default.
• W2954781872 cites W21025885 @default.
• W2954781872 cites W2128340050 @default.
• W2954781872 cites W2156777442 @default.
• W2954781872 cites W2157554677 @default.
• W2954781872 cites W2161236525 @default.
• W2954781872 cites W2162681317 @default.
• W2954781872 cites W2166650627 @default.
• W2954781872 cites W2194775991 @default.
• W2954781872 cites W2293332611 @default.
• W2954781872 cites W2338972621 @default.
• W2954781872 cites W2400000673 @default.
• W2954781872 cites W2412782625 @default.
• W2954781872 cites W2437041077 @default.
• W2954781872 cites W2461475918 @default.
• W2954781872 cites W2470270897 @default.
• W2954781872 cites W2470948946 @default.
• W2954781872 cites W2519623608 @default.
• W2954781872 cites W2549139847 @default.
• W2954781872 cites W2560023338 @default.
• W2954781872 cites W2565639579 @default.
• W2954781872 cites W2567978322 @default.
• W2954781872 cites W2569272946 @default.
• W2954781872 cites W2571295082 @default.
• W2954781872 cites W2605929543 @default.
• W2954781872 cites W2613718673 @default.
• W2954781872 cites W2615461745 @default.
• W2954781872 cites W2630837129 @default.
• W2954781872 cites W2740652190 @default.
• W2954781872 cites W2740667773 @default.
• W2954781872 cites W2744613561 @default.
• W2954781872 cites W2754188632 @default.
• W2954781872 cites W2772161954 @default.
• W2954781872 cites W2777511827 @default.
• W2954781872 cites W2777654136 @default.
• W2954781872 cites W2780708736 @default.
• W2954781872 cites W2792873273 @default.
• W2954781872 cites W2795251047 @default.
• W2954781872 cites W2798355657 @default.
• W2954781872 cites W2798791651 @default.
• W2954781872 cites W2798807298 @default.
• W2954781872 cites W2798857366 @default.
• W2954781872 cites W2799074129 @default.
• W2954781872 cites W2807746031 @default.
• W2954781872 cites W2884555738 @default.
• W2954781872 cites W2894666165 @default.
• W2954781872 cites W2895251968 @default.
• W2954781872 cites W2934493198 @default.
• W2954781872 cites W2939217524 @default.
• W2954781872 cites W2962835968 @default.
• W2954781872 cites W2963014378 @default.
• W2954781872 cites W2963032190 @default.
• W2954781872 cites W2963057024 @default.
• W2954781872 cites W2963112696 @default.
• W2954781872 cites W2963150697 @default.
• W2954781872 cites W2963334022 @default.
• W2954781872 cites W2963342610 @default.
• W2954781872 cites W2963635628 @default.
• W2954781872 cites W2963685207 @default.
• W2954781872 cites W2963857746 @default.
• W2954781872 cites W2963906836 @default.
• W2954781872 cites W2963915286 @default.
• W2954781872 cites W2963947444 @default.
• W2954781872 cites W2964236837 @default.
• W2954781872 cites W2964352379 @default.
• W2954781872 cites W2986825110 @default.
• W2954781872 cites W3098389804 @default.
• W2954781872 cites W3125520697 @default.
• W2954781872 cites W3127842933 @default.
• W2954781872 cites W845365781 @default.
• W2954781872 doi "https://doi.org/10.48550/arxiv.1906.11443" @default.
• W2954781872 hasPublicationYear "2019" @default.
• W2954781872 type Work @default.
• W2954781872 sameAs 2954781872 @default.
• W2954781872 citedByCount "0" @default.
• W2954781872 crossrefType "posted-content" @default.

• next