Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386066405> ?p ?o ?g. }
- W4386066405 abstract "Learning to leverage the relationship among diverse image restoration tasks is quite beneficial for unraveling the intrinsicingredients behind the degradation. Recent years have witnessed the flourish of various All-in-one methods, which handle multiple image degradations within a single model. In practice, however, few attempts have been made to excavate task correlations in that exploring the underlying fundamentalingredients of various image degradations, resulting in poor scalability as more tasks are involved. In this paper, we propose a novel perspective to delve into the degradation via aningredients-oriented rather than previous task-oriented manner for scalable learning. Specifically, our method, named Ingredients-oriented Degradation Reformulation framework (IDR), consists of two stages, namely task-oriented knowledge collection and ingredients-oriented knowledge integration. In the first stage, we conduct ad hoc operations on different degradations according to the underlying physics principles, and establish the corresponding prior hubs for each type of degradation. While the second stage progressively reformulates the preceding task-oriented hubs into single ingredients-oriented hub via learnable Principal Component Analysis (PCA), and employs a dynamic routing mechanism for probabilistic unknown degradation removal. Extensive experiments on various image restoration tasks demonstrate the effectiveness and scalability of our method. More importantly, our IDR exhibits the favorable generalization ability to unknown downstream tasks." @default.
- W4386066405 created "2023-08-23" @default.
- W4386066405 creator A5002617843 @default.
- W4386066405 creator A5005310183 @default.
- W4386066405 creator A5024161331 @default.
- W4386066405 creator A5026178309 @default.
- W4386066405 creator A5033737940 @default.
- W4386066405 creator A5052274720 @default.
- W4386066405 creator A5062166339 @default.
- W4386066405 date "2023-06-01" @default.
- W4386066405 modified "2023-09-27" @default.
- W4386066405 title "Ingredient-oriented Multi-Degradation Learning for Image Restoration" @default.
- W4386066405 cites W1499332833 @default.
- W4386066405 cites W1930824406 @default.
- W4386066405 cites W2056370875 @default.
- W4386066405 cites W2063784584 @default.
- W4386066405 cites W2076205488 @default.
- W4386066405 cites W2121880036 @default.
- W4386066405 cites W2121927366 @default.
- W4386066405 cites W2556068545 @default.
- W4386066405 cites W2559264300 @default.
- W4386066405 cites W2560533888 @default.
- W4386066405 cites W2779176852 @default.
- W4386066405 cites W2891076394 @default.
- W4386066405 cites W2913340405 @default.
- W4386066405 cites W2962785568 @default.
- W4386066405 cites W2963249268 @default.
- W4386066405 cites W2963430933 @default.
- W4386066405 cites W2963641969 @default.
- W4386066405 cites W2963725279 @default.
- W4386066405 cites W2963877604 @default.
- W4386066405 cites W2963928582 @default.
- W4386066405 cites W2998249728 @default.
- W4386066405 cites W3035086889 @default.
- W4386066405 cites W3035250394 @default.
- W4386066405 cites W3035326127 @default.
- W4386066405 cites W3035713416 @default.
- W4386066405 cites W3035731588 @default.
- W4386066405 cites W3171125843 @default.
- W4386066405 cites W3175634226 @default.
- W4386066405 cites W3176096490 @default.
- W4386066405 cites W3177235959 @default.
- W4386066405 cites W3181566645 @default.
- W4386066405 cites W3202040256 @default.
- W4386066405 cites W3207918547 @default.
- W4386066405 cites W4226440503 @default.
- W4386066405 cites W4226453196 @default.
- W4386066405 cites W4242059867 @default.
- W4386066405 cites W4283023197 @default.
- W4386066405 cites W4312249431 @default.
- W4386066405 cites W4312373367 @default.
- W4386066405 cites W4312575110 @default.
- W4386066405 cites W4312725970 @default.
- W4386066405 cites W4312812783 @default.
- W4386066405 cites W4312908055 @default.
- W4386066405 doi "https://doi.org/10.1109/cvpr52729.2023.00564" @default.
- W4386066405 hasPublicationYear "2023" @default.
- W4386066405 type Work @default.
- W4386066405 citedByCount "0" @default.
- W4386066405 crossrefType "proceedings-article" @default.
- W4386066405 hasAuthorship W4386066405A5002617843 @default.
- W4386066405 hasAuthorship W4386066405A5005310183 @default.
- W4386066405 hasAuthorship W4386066405A5024161331 @default.
- W4386066405 hasAuthorship W4386066405A5026178309 @default.
- W4386066405 hasAuthorship W4386066405A5033737940 @default.
- W4386066405 hasAuthorship W4386066405A5052274720 @default.
- W4386066405 hasAuthorship W4386066405A5062166339 @default.
- W4386066405 hasConcept C115961682 @default.
- W4386066405 hasConcept C119857082 @default.
- W4386066405 hasConcept C12713177 @default.
- W4386066405 hasConcept C127413603 @default.
- W4386066405 hasConcept C134306372 @default.
- W4386066405 hasConcept C153083717 @default.
- W4386066405 hasConcept C154945302 @default.
- W4386066405 hasConcept C177148314 @default.
- W4386066405 hasConcept C201995342 @default.
- W4386066405 hasConcept C2779679103 @default.
- W4386066405 hasConcept C2780451532 @default.
- W4386066405 hasConcept C33923547 @default.
- W4386066405 hasConcept C41008148 @default.
- W4386066405 hasConcept C48044578 @default.
- W4386066405 hasConcept C49937458 @default.
- W4386066405 hasConcept C76155785 @default.
- W4386066405 hasConcept C77088390 @default.
- W4386066405 hasConceptScore W4386066405C115961682 @default.
- W4386066405 hasConceptScore W4386066405C119857082 @default.
- W4386066405 hasConceptScore W4386066405C12713177 @default.
- W4386066405 hasConceptScore W4386066405C127413603 @default.
- W4386066405 hasConceptScore W4386066405C134306372 @default.
- W4386066405 hasConceptScore W4386066405C153083717 @default.
- W4386066405 hasConceptScore W4386066405C154945302 @default.
- W4386066405 hasConceptScore W4386066405C177148314 @default.
- W4386066405 hasConceptScore W4386066405C201995342 @default.
- W4386066405 hasConceptScore W4386066405C2779679103 @default.
- W4386066405 hasConceptScore W4386066405C2780451532 @default.
- W4386066405 hasConceptScore W4386066405C33923547 @default.
- W4386066405 hasConceptScore W4386066405C41008148 @default.
- W4386066405 hasConceptScore W4386066405C48044578 @default.
- W4386066405 hasConceptScore W4386066405C49937458 @default.
- W4386066405 hasConceptScore W4386066405C76155785 @default.