FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3097337310> ?p ?o ?g. }

• W3097337310 endingPage "435" @default.
• W3097337310 startingPage "419" @default.
• W3097337310 abstract "Visual saliency modeling for images and videos is treated as two independent tasks in recent computer vision literature. While image saliency modeling is a well-studied problem and progress on benchmarks like SALICON and MIT300 is slowing, video saliency models have shown rapid gains on the recent DHF1K benchmark. Here, we take a step back and ask: Can image and video saliency modeling be approached via a unified model, with mutual benefit? We identify different sources of domain shift between image and video saliency data and between different video saliency datasets as a key challenge for effective joint modelling. To address this we propose four novel domain adaptation techniques - Domain-Adaptive Priors, Domain-Adaptive Fusion, Domain-Adaptive Smoothing and Bypass-RNN - in addition to an improved formulation of learned Gaussian priors. We integrate these techniques into a simple and lightweight encoder-RNN-decoder-style network, UNISAL, and train it jointly with image and video saliency data. We evaluate our method on the video saliency datasets DHF1K, Hollywood-2 and UCF-Sports, and the image saliency datasets SALICON and MIT300. With one set of parameters, UNISAL achieves state-of-the-art performance on all video saliency datasets and is on par with the state-of-the-art for image saliency datasets, despite faster runtime and a 5 to 20-fold smaller model size compared to all competing deep methods. We provide retrospective analyses and ablation studies which confirm the importance of the domain shift modeling. The code is available at https://github.com/rdroste/unisal" @default.
• W3097337310 created "2020-11-09" @default.
• W3097337310 creator A5017599481 @default.
• W3097337310 creator A5039629184 @default.
• W3097337310 creator A5042467582 @default.
• W3097337310 date "2020-01-01" @default.
• W3097337310 modified "2023-10-17" @default.
• W3097337310 title "Unified Image and Video Saliency Modeling" @default.
• W3097337310 cites W1510835000 @default.
• W3097337310 cites W1934890906 @default.
• W3097337310 cites W1959094031 @default.
• W3097337310 cites W1970342316 @default.
• W3097337310 cites W1987794317 @default.
• W3097337310 cites W2033859430 @default.
• W3097337310 cites W2034436892 @default.
• W3097337310 cites W2071555787 @default.
• W3097337310 cites W2078903912 @default.
• W3097337310 cites W2128272608 @default.
• W3097337310 cites W2162216530 @default.
• W3097337310 cites W2164084182 @default.
• W3097337310 cites W2164720308 @default.
• W3097337310 cites W2212216676 @default.
• W3097337310 cites W2288514685 @default.
• W3097337310 cites W2295598507 @default.
• W3097337310 cites W2312004824 @default.
• W3097337310 cites W2342611082 @default.
• W3097337310 cites W2378845821 @default.
• W3097337310 cites W2442293398 @default.
• W3097337310 cites W2510185399 @default.
• W3097337310 cites W2753641513 @default.
• W3097337310 cites W2775654873 @default.
• W3097337310 cites W2799077596 @default.
• W3097337310 cites W2896172606 @default.
• W3097337310 cites W2949813473 @default.
• W3097337310 cites W2962965915 @default.
• W3097337310 cites W2963163009 @default.
• W3097337310 cites W2963503775 @default.
• W3097337310 cites W2963631529 @default.
• W3097337310 cites W2980565715 @default.
• W3097337310 cites W2986131415 @default.
• W3097337310 cites W3099561715 @default.
• W3097337310 doi "https://doi.org/10.1007/978-3-030-58558-7_25" @default.
• W3097337310 hasPublicationYear "2020" @default.
• W3097337310 type Work @default.
• W3097337310 sameAs 3097337310 @default.
• W3097337310 citedByCount "61" @default.
• W3097337310 countsByYear W30973373102020 @default.
• W3097337310 countsByYear W30973373102021 @default.
• W3097337310 countsByYear W30973373102022 @default.
• W3097337310 countsByYear W30973373102023 @default.
• W3097337310 crossrefType "book-chapter" @default.
• W3097337310 hasAuthorship W3097337310A5017599481 @default.
• W3097337310 hasAuthorship W3097337310A5039629184 @default.
• W3097337310 hasAuthorship W3097337310A5042467582 @default.
• W3097337310 hasBestOaLocation W30973373102 @default.
• W3097337310 hasConcept C111919701 @default.
• W3097337310 hasConcept C115961682 @default.
• W3097337310 hasConcept C118505674 @default.
• W3097337310 hasConcept C13280743 @default.
• W3097337310 hasConcept C134306372 @default.
• W3097337310 hasConcept C153180895 @default.
• W3097337310 hasConcept C154945302 @default.
• W3097337310 hasConcept C185798385 @default.
• W3097337310 hasConcept C205649164 @default.
• W3097337310 hasConcept C2780580889 @default.
• W3097337310 hasConcept C31972630 @default.
• W3097337310 hasConcept C33923547 @default.
• W3097337310 hasConcept C36503486 @default.
• W3097337310 hasConcept C3770464 @default.
• W3097337310 hasConcept C41008148 @default.
• W3097337310 hasConceptScore W3097337310C111919701 @default.
• W3097337310 hasConceptScore W3097337310C115961682 @default.
• W3097337310 hasConceptScore W3097337310C118505674 @default.
• W3097337310 hasConceptScore W3097337310C13280743 @default.
• W3097337310 hasConceptScore W3097337310C134306372 @default.
• W3097337310 hasConceptScore W3097337310C153180895 @default.
• W3097337310 hasConceptScore W3097337310C154945302 @default.
• W3097337310 hasConceptScore W3097337310C185798385 @default.
• W3097337310 hasConceptScore W3097337310C205649164 @default.
• W3097337310 hasConceptScore W3097337310C2780580889 @default.
• W3097337310 hasConceptScore W3097337310C31972630 @default.
• W3097337310 hasConceptScore W3097337310C33923547 @default.
• W3097337310 hasConceptScore W3097337310C36503486 @default.
• W3097337310 hasConceptScore W3097337310C3770464 @default.
• W3097337310 hasConceptScore W3097337310C41008148 @default.
• W3097337310 hasLocation W30973373101 @default.
• W3097337310 hasLocation W30973373102 @default.
• W3097337310 hasLocation W30973373103 @default.
• W3097337310 hasOpenAccess W3097337310 @default.
• W3097337310 hasPrimaryLocation W30973373101 @default.
• W3097337310 hasRelatedWork W2071240554 @default.
• W3097337310 hasRelatedWork W2081023556 @default.
• W3097337310 hasRelatedWork W2122348443 @default.
• W3097337310 hasRelatedWork W2152245267 @default.
• W3097337310 hasRelatedWork W2167361453 @default.
• W3097337310 hasRelatedWork W2169995962 @default.
• W3097337310 hasRelatedWork W2771862904 @default.
• W3097337310 hasRelatedWork W3010087653 @default.

• next