Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386590405> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W4386590405 abstract "The limitation of the Human Visual System (HVS) in perceiving small distortions allows us to lower the bitrate required to achieve a certain visual quality. Predicting and applying the Just Noticeable Distortion (JND), which is a threshold for maximum unperceived level of distortions, is among the popular ways to do so. Recently, machine learning based methods have been able to reduce bitrate even further by improving JND prediction accuracy. However, accurate modeling of JND is very challenging, as it is highly content dependent. Furthermore, existing datasets provide little information to learn the best parameters. To remedy this issue, we propose a multi-task deep learning framework that jointly learns various complementary visual information. We design three separate methods and training strategies that jointly learn: (1) three JND levels, (2) visual attention map and a JND level, and (3) three JND levels and the visual attention map. We show that accumulating information from multiple tasks leads to a more robust prediction of JND. Experimental results confirm the superiority of our framework compared to the state-of-the-art." @default.
- W4386590405 created "2023-09-11" @default.
- W4386590405 creator A5005099415 @default.
- W4386590405 creator A5023878386 @default.
- W4386590405 creator A5036640771 @default.
- W4386590405 creator A5044422170 @default.
- W4386590405 creator A5072482381 @default.
- W4386590405 date "2023-10-08" @default.
- W4386590405 modified "2023-09-30" @default.
- W4386590405 title "MTJND: Multi-Task Deep Learning Framework for Improved JND Prediction" @default.
- W4386590405 cites W2117539524 @default.
- W4386590405 cites W2147347517 @default.
- W4386590405 cites W2321545007 @default.
- W4386590405 cites W2386401927 @default.
- W4386590405 cites W2511458122 @default.
- W4386590405 cites W2575940230 @default.
- W4386590405 cites W2581531361 @default.
- W4386590405 cites W2890640783 @default.
- W4386590405 cites W2962785568 @default.
- W4386590405 cites W2968674715 @default.
- W4386590405 cites W2971351108 @default.
- W4386590405 cites W3012085781 @default.
- W4386590405 cites W3034217182 @default.
- W4386590405 cites W3046565475 @default.
- W4386590405 cites W3095429212 @default.
- W4386590405 cites W3099561715 @default.
- W4386590405 cites W3100498948 @default.
- W4386590405 cites W3105798550 @default.
- W4386590405 cites W3120556731 @default.
- W4386590405 cites W3126164993 @default.
- W4386590405 cites W3141797743 @default.
- W4386590405 cites W3159346357 @default.
- W4386590405 cites W3189792606 @default.
- W4386590405 cites W4286655730 @default.
- W4386590405 doi "https://doi.org/10.1109/icip49359.2023.10222099" @default.
- W4386590405 hasPublicationYear "2023" @default.
- W4386590405 type Work @default.
- W4386590405 citedByCount "0" @default.
- W4386590405 crossrefType "proceedings-article" @default.
- W4386590405 hasAuthorship W4386590405A5005099415 @default.
- W4386590405 hasAuthorship W4386590405A5023878386 @default.
- W4386590405 hasAuthorship W4386590405A5036640771 @default.
- W4386590405 hasAuthorship W4386590405A5044422170 @default.
- W4386590405 hasAuthorship W4386590405A5072482381 @default.
- W4386590405 hasBestOaLocation W43865904051 @default.
- W4386590405 hasConcept C108583219 @default.
- W4386590405 hasConcept C115961682 @default.
- W4386590405 hasConcept C119857082 @default.
- W4386590405 hasConcept C126780896 @default.
- W4386590405 hasConcept C153180895 @default.
- W4386590405 hasConcept C154945302 @default.
- W4386590405 hasConcept C160086991 @default.
- W4386590405 hasConcept C162324750 @default.
- W4386590405 hasConcept C187736073 @default.
- W4386590405 hasConcept C194257627 @default.
- W4386590405 hasConcept C2776257435 @default.
- W4386590405 hasConcept C2780451532 @default.
- W4386590405 hasConcept C31258907 @default.
- W4386590405 hasConcept C36464697 @default.
- W4386590405 hasConcept C41008148 @default.
- W4386590405 hasConcept C59111918 @default.
- W4386590405 hasConceptScore W4386590405C108583219 @default.
- W4386590405 hasConceptScore W4386590405C115961682 @default.
- W4386590405 hasConceptScore W4386590405C119857082 @default.
- W4386590405 hasConceptScore W4386590405C126780896 @default.
- W4386590405 hasConceptScore W4386590405C153180895 @default.
- W4386590405 hasConceptScore W4386590405C154945302 @default.
- W4386590405 hasConceptScore W4386590405C160086991 @default.
- W4386590405 hasConceptScore W4386590405C162324750 @default.
- W4386590405 hasConceptScore W4386590405C187736073 @default.
- W4386590405 hasConceptScore W4386590405C194257627 @default.
- W4386590405 hasConceptScore W4386590405C2776257435 @default.
- W4386590405 hasConceptScore W4386590405C2780451532 @default.
- W4386590405 hasConceptScore W4386590405C31258907 @default.
- W4386590405 hasConceptScore W4386590405C36464697 @default.
- W4386590405 hasConceptScore W4386590405C41008148 @default.
- W4386590405 hasConceptScore W4386590405C59111918 @default.
- W4386590405 hasFunder F4320328501 @default.
- W4386590405 hasLocation W43865904051 @default.
- W4386590405 hasOpenAccess W4386590405 @default.
- W4386590405 hasPrimaryLocation W43865904051 @default.
- W4386590405 hasRelatedWork W2795261237 @default.
- W4386590405 hasRelatedWork W3014300295 @default.
- W4386590405 hasRelatedWork W3164822677 @default.
- W4386590405 hasRelatedWork W4223943233 @default.
- W4386590405 hasRelatedWork W4225161397 @default.
- W4386590405 hasRelatedWork W4312200629 @default.
- W4386590405 hasRelatedWork W4360585206 @default.
- W4386590405 hasRelatedWork W4364306694 @default.
- W4386590405 hasRelatedWork W4380075502 @default.
- W4386590405 hasRelatedWork W4380086463 @default.
- W4386590405 isParatext "false" @default.
- W4386590405 isRetracted "false" @default.
- W4386590405 workType "article" @default.