Matches in SemOpenAlex for { <https://semopenalex.org/work/W3036523977> ?p ?o ?g. }
- W3036523977 abstract "Lossy image compression is one of the most commonly used operators for digital images. Most recently proposed deep-learning-based image compression methods leverage the auto-encoder structure, and reach a series of promising results in this field. The images are encoded into low dimensional latent features first, and entropy coded subsequently by exploiting the statistical redundancy. However, the information lost during encoding is unfortunately inevitable, which poses a significant challenge to the decoder to reconstruct the original images. In this work, we propose a novel invertible framework called Invertible Lossy Compression (ILC) to largely mitigate the information loss problem. Specifically, ILC introduces an invertible encoding module to replace the encoder-decoder structure to produce the low dimensional informative latent representation, meanwhile, transform the lost information into an auxiliary latent variable that won't be further coded or stored. The latent representation is quantized and encoded into bit-stream, and the latent variable is forced to follow a specified distribution, i.e. isotropic Gaussian distribution. In this way, recovering the original image is made tractable by easily drawing a surrogate latent variable and applying the inverse pass of the module with the sampled variable and decoded latent features. Experimental results demonstrate that with a new component replacing the auto-encoder in image compression methods, ILC can significantly outperform the baseline method on extensive benchmark datasets by combining with the existing compression algorithms." @default.
- W3036523977 created "2020-06-25" @default.
- W3036523977 creator A5022256556 @default.
- W3036523977 creator A5051093347 @default.
- W3036523977 creator A5064786542 @default.
- W3036523977 creator A5070990160 @default.
- W3036523977 creator A5075482809 @default.
- W3036523977 date "2020-06-22" @default.
- W3036523977 modified "2023-09-28" @default.
- W3036523977 title "Modeling Lost Information in Lossy Image Compression." @default.
- W3036523977 cites W1522301498 @default.
- W3036523977 cites W1821462560 @default.
- W3036523977 cites W2012762214 @default.
- W3036523977 cites W2108598243 @default.
- W3036523977 cites W2118217749 @default.
- W3036523977 cites W2119047110 @default.
- W3036523977 cites W2134797427 @default.
- W3036523977 cites W2140196014 @default.
- W3036523977 cites W2148593155 @default.
- W3036523977 cites W2242818861 @default.
- W3036523977 cites W2294370754 @default.
- W3036523977 cites W2887540873 @default.
- W3036523977 cites W2892278106 @default.
- W3036523977 cites W2893920456 @default.
- W3036523977 cites W2935381027 @default.
- W3036523977 cites W2938908220 @default.
- W3036523977 cites W2949995983 @default.
- W3036523977 cites W2954340342 @default.
- W3036523977 cites W2962676454 @default.
- W3036523977 cites W2962695743 @default.
- W3036523977 cites W2963066125 @default.
- W3036523977 cites W2963139417 @default.
- W3036523977 cites W2963684275 @default.
- W3036523977 cites W2963711615 @default.
- W3036523977 cites W2964098744 @default.
- W3036523977 cites W2969569226 @default.
- W3036523977 cites W2981613960 @default.
- W3036523977 cites W2996631524 @default.
- W3036523977 cites W3023717151 @default.
- W3036523977 cites W3025996576 @default.
- W3036523977 cites W80079294 @default.
- W3036523977 hasPublicationYear "2020" @default.
- W3036523977 type Work @default.
- W3036523977 sameAs 3036523977 @default.
- W3036523977 citedByCount "1" @default.
- W3036523977 countsByYear W30365239772021 @default.
- W3036523977 crossrefType "posted-content" @default.
- W3036523977 hasAuthorship W3036523977A5022256556 @default.
- W3036523977 hasAuthorship W3036523977A5051093347 @default.
- W3036523977 hasAuthorship W3036523977A5064786542 @default.
- W3036523977 hasAuthorship W3036523977A5070990160 @default.
- W3036523977 hasAuthorship W3036523977A5075482809 @default.
- W3036523977 hasConcept C111919701 @default.
- W3036523977 hasConcept C11413529 @default.
- W3036523977 hasConcept C115961682 @default.
- W3036523977 hasConcept C118505674 @default.
- W3036523977 hasConcept C13481523 @default.
- W3036523977 hasConcept C153180895 @default.
- W3036523977 hasConcept C154945302 @default.
- W3036523977 hasConcept C165021410 @default.
- W3036523977 hasConcept C31972630 @default.
- W3036523977 hasConcept C41008148 @default.
- W3036523977 hasConcept C51167844 @default.
- W3036523977 hasConcept C78548338 @default.
- W3036523977 hasConcept C80444323 @default.
- W3036523977 hasConcept C9417928 @default.
- W3036523977 hasConceptScore W3036523977C111919701 @default.
- W3036523977 hasConceptScore W3036523977C11413529 @default.
- W3036523977 hasConceptScore W3036523977C115961682 @default.
- W3036523977 hasConceptScore W3036523977C118505674 @default.
- W3036523977 hasConceptScore W3036523977C13481523 @default.
- W3036523977 hasConceptScore W3036523977C153180895 @default.
- W3036523977 hasConceptScore W3036523977C154945302 @default.
- W3036523977 hasConceptScore W3036523977C165021410 @default.
- W3036523977 hasConceptScore W3036523977C31972630 @default.
- W3036523977 hasConceptScore W3036523977C41008148 @default.
- W3036523977 hasConceptScore W3036523977C51167844 @default.
- W3036523977 hasConceptScore W3036523977C78548338 @default.
- W3036523977 hasConceptScore W3036523977C80444323 @default.
- W3036523977 hasConceptScore W3036523977C9417928 @default.
- W3036523977 hasOpenAccess W3036523977 @default.
- W3036523977 hasRelatedWork W1544418007 @default.
- W3036523977 hasRelatedWork W2095987654 @default.
- W3036523977 hasRelatedWork W2097381298 @default.
- W3036523977 hasRelatedWork W2183728785 @default.
- W3036523977 hasRelatedWork W2340721861 @default.
- W3036523977 hasRelatedWork W2907543445 @default.
- W3036523977 hasRelatedWork W2997673699 @default.
- W3036523977 hasRelatedWork W3014578089 @default.
- W3036523977 hasRelatedWork W3021004169 @default.
- W3036523977 hasRelatedWork W3103807521 @default.
- W3036523977 hasRelatedWork W3111906665 @default.
- W3036523977 hasRelatedWork W3175247443 @default.
- W3036523977 hasRelatedWork W3176485436 @default.
- W3036523977 hasRelatedWork W3202779111 @default.
- W3036523977 hasRelatedWork W3210965296 @default.
- W3036523977 hasRelatedWork W3213713113 @default.
- W3036523977 hasRelatedWork W60670607 @default.
- W3036523977 hasRelatedWork W945687790 @default.
- W3036523977 hasRelatedWork W2185078603 @default.