Matches in SemOpenAlex for { <https://semopenalex.org/work/W3132232437> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3132232437 endingPage "116201" @default.
- W3132232437 startingPage "116201" @default.
- W3132232437 abstract "Convolutional Neural Network (CNN) structures have been designed for in-loop filtering to improve video coding performance. These CNN models are usually trained through learning the correlations between the reconstructed and the original frames, which are then applied to every single reconstructed frame to improve the overall video quality. This direct model training and deployment strategy is effective for intra coding since a locally optimal model is sufficient. However, when applied to inter coding, it causes over-filtering because the intertwined reference dependencies across inter frames are not taken into consideration. To address this issue, existing methods usually resort to the Rate–Distortion Optimization (RDO) to selectively apply the CNN model, but fail to address the limitation of using a local CNN model. In this paper, we propose a progressive approach to train and incorporate the CNN-based in-loop filters to work seamlessly with video encoders. First, we develop a progressive training method to obtain the inter model. Using transfer learning, reconstructed frames using the CNN model are progressively involved back into the training of the CNN model itself, to simulate the reference dependencies in inter coding. Next, we design a frame-level model selection strategy for the high-bitrate coding where the over-filtering effect is diluted. Experimental results show that the proposed method outperforms the RDO method that utilizes only local model. Proposed approach also achieves comparable coding performance but with less computational complexity when integrating our progressive model into the RDO scheme." @default.
- W3132232437 created "2021-03-01" @default.
- W3132232437 creator A5013303786 @default.
- W3132232437 creator A5014275265 @default.
- W3132232437 creator A5028059517 @default.
- W3132232437 creator A5061501836 @default.
- W3132232437 date "2021-05-01" @default.
- W3132232437 modified "2023-10-03" @default.
- W3132232437 title "A progressive CNN in-loop filtering approach for inter frame coding" @default.
- W3132232437 cites W2043292349 @default.
- W3132232437 cites W2142683286 @default.
- W3132232437 cites W2242218935 @default.
- W3132232437 cites W2278490765 @default.
- W3132232437 cites W2477177239 @default.
- W3132232437 cites W2508457857 @default.
- W3132232437 cites W2510648513 @default.
- W3132232437 cites W2519021537 @default.
- W3132232437 cites W2612222456 @default.
- W3132232437 cites W2741137940 @default.
- W3132232437 cites W2752154940 @default.
- W3132232437 cites W2753574358 @default.
- W3132232437 cites W2791152551 @default.
- W3132232437 cites W2792186562 @default.
- W3132232437 cites W2792275277 @default.
- W3132232437 cites W2793408333 @default.
- W3132232437 cites W2884796644 @default.
- W3132232437 cites W2890409760 @default.
- W3132232437 cites W2890577598 @default.
- W3132232437 cites W2891639355 @default.
- W3132232437 cites W2913664580 @default.
- W3132232437 cites W2941995653 @default.
- W3132232437 cites W2964157487 @default.
- W3132232437 cites W2970384285 @default.
- W3132232437 cites W2970450539 @default.
- W3132232437 cites W2970669588 @default.
- W3132232437 cites W3000128039 @default.
- W3132232437 cites W3000502695 @default.
- W3132232437 cites W3100087914 @default.
- W3132232437 cites W3100870787 @default.
- W3132232437 cites W3102974666 @default.
- W3132232437 cites W4239851455 @default.
- W3132232437 doi "https://doi.org/10.1016/j.image.2021.116201" @default.
- W3132232437 hasPublicationYear "2021" @default.
- W3132232437 type Work @default.
- W3132232437 sameAs 3132232437 @default.
- W3132232437 citedByCount "4" @default.
- W3132232437 countsByYear W31322324372021 @default.
- W3132232437 countsByYear W31322324372022 @default.
- W3132232437 crossrefType "journal-article" @default.
- W3132232437 hasAuthorship W3132232437A5013303786 @default.
- W3132232437 hasAuthorship W3132232437A5014275265 @default.
- W3132232437 hasAuthorship W3132232437A5028059517 @default.
- W3132232437 hasAuthorship W3132232437A5061501836 @default.
- W3132232437 hasConcept C105795698 @default.
- W3132232437 hasConcept C111919701 @default.
- W3132232437 hasConcept C11413529 @default.
- W3132232437 hasConcept C116709606 @default.
- W3132232437 hasConcept C118505674 @default.
- W3132232437 hasConcept C153180895 @default.
- W3132232437 hasConcept C154945302 @default.
- W3132232437 hasConcept C179518139 @default.
- W3132232437 hasConcept C31972630 @default.
- W3132232437 hasConcept C33923547 @default.
- W3132232437 hasConcept C41008148 @default.
- W3132232437 hasConcept C81363708 @default.
- W3132232437 hasConceptScore W3132232437C105795698 @default.
- W3132232437 hasConceptScore W3132232437C111919701 @default.
- W3132232437 hasConceptScore W3132232437C11413529 @default.
- W3132232437 hasConceptScore W3132232437C116709606 @default.
- W3132232437 hasConceptScore W3132232437C118505674 @default.
- W3132232437 hasConceptScore W3132232437C153180895 @default.
- W3132232437 hasConceptScore W3132232437C154945302 @default.
- W3132232437 hasConceptScore W3132232437C179518139 @default.
- W3132232437 hasConceptScore W3132232437C31972630 @default.
- W3132232437 hasConceptScore W3132232437C33923547 @default.
- W3132232437 hasConceptScore W3132232437C41008148 @default.
- W3132232437 hasConceptScore W3132232437C81363708 @default.
- W3132232437 hasLocation W31322324371 @default.
- W3132232437 hasOpenAccess W3132232437 @default.
- W3132232437 hasPrimaryLocation W31322324371 @default.
- W3132232437 hasRelatedWork W2015942816 @default.
- W3132232437 hasRelatedWork W2151453963 @default.
- W3132232437 hasRelatedWork W2170630590 @default.
- W3132232437 hasRelatedWork W2175746458 @default.
- W3132232437 hasRelatedWork W2732542196 @default.
- W3132232437 hasRelatedWork W2738221750 @default.
- W3132232437 hasRelatedWork W2760085659 @default.
- W3132232437 hasRelatedWork W2883200793 @default.
- W3132232437 hasRelatedWork W3012978760 @default.
- W3132232437 hasRelatedWork W3093612317 @default.
- W3132232437 hasVolume "94" @default.
- W3132232437 isParatext "false" @default.
- W3132232437 isRetracted "false" @default.
- W3132232437 magId "3132232437" @default.
- W3132232437 workType "article" @default.