Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226168429> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W4226168429 abstract "This work addresses two major issues of end-to-end learned image compression (LIC) based on deep neural networks: variable-rate learning where separate networks are required to generate compressed images with varying qualities, and the train-test mismatch between differentiable approximate quantization and true hard quantization. We introduce an online meta-learning (OML) setting for LIC, which combines ideas from meta learning and online learning in the conditional variational auto-encoder (CVAE) framework. By treating the conditional variables as meta parameters and treating the generated conditional features as meta priors, the desired reconstruction can be controlled by the meta parameters to accommodate compression with variable qualities. The online learning framework is used to update the meta parameters so that the conditional reconstruction is adaptively tuned for the current image. Through the OML mechanism, the meta parameters can be effectively updated through SGD. The conditional reconstruction is directly based on the quantized latent representation in the decoder network, and therefore helps to bridge the gap between the training estimation and true quantized latent distribution. Experiments demonstrate that our OML approach can be flexibly applied to different state-of-the-art LIC methods to achieve additional performance improvements with little computation and transmission overhead." @default.
- W4226168429 created "2022-05-05" @default.
- W4226168429 creator A5002942261 @default.
- W4226168429 creator A5046597133 @default.
- W4226168429 creator A5082496274 @default.
- W4226168429 creator A5083218433 @default.
- W4226168429 date "2022-06-01" @default.
- W4226168429 modified "2023-09-29" @default.
- W4226168429 title "Online Meta Adaptation for Variable-Rate Learned Image Compression" @default.
- W4226168429 cites W2030265202 @default.
- W4226168429 cites W2040903332 @default.
- W4226168429 cites W2146395539 @default.
- W4226168429 cites W2962891349 @default.
- W4226168429 cites W2963149687 @default.
- W4226168429 cites W2963739929 @default.
- W4226168429 cites W2982853315 @default.
- W4226168429 cites W3034469748 @default.
- W4226168429 cites W3035304632 @default.
- W4226168429 cites W3104308854 @default.
- W4226168429 cites W3147173580 @default.
- W4226168429 cites W3175457126 @default.
- W4226168429 cites W3183453506 @default.
- W4226168429 doi "https://doi.org/10.1109/cvprw56347.2022.00065" @default.
- W4226168429 hasPublicationYear "2022" @default.
- W4226168429 type Work @default.
- W4226168429 citedByCount "3" @default.
- W4226168429 countsByYear W42261684292023 @default.
- W4226168429 crossrefType "proceedings-article" @default.
- W4226168429 hasAuthorship W4226168429A5002942261 @default.
- W4226168429 hasAuthorship W4226168429A5046597133 @default.
- W4226168429 hasAuthorship W4226168429A5082496274 @default.
- W4226168429 hasAuthorship W4226168429A5083218433 @default.
- W4226168429 hasBestOaLocation W42261684292 @default.
- W4226168429 hasConcept C107673813 @default.
- W4226168429 hasConcept C11413529 @default.
- W4226168429 hasConcept C115961682 @default.
- W4226168429 hasConcept C119857082 @default.
- W4226168429 hasConcept C13481523 @default.
- W4226168429 hasConcept C154945302 @default.
- W4226168429 hasConcept C162324750 @default.
- W4226168429 hasConcept C177769412 @default.
- W4226168429 hasConcept C187736073 @default.
- W4226168429 hasConcept C2780451532 @default.
- W4226168429 hasConcept C2781002164 @default.
- W4226168429 hasConcept C28855332 @default.
- W4226168429 hasConcept C41008148 @default.
- W4226168429 hasConcept C51167844 @default.
- W4226168429 hasConcept C9417928 @default.
- W4226168429 hasConcept C94835093 @default.
- W4226168429 hasConceptScore W4226168429C107673813 @default.
- W4226168429 hasConceptScore W4226168429C11413529 @default.
- W4226168429 hasConceptScore W4226168429C115961682 @default.
- W4226168429 hasConceptScore W4226168429C119857082 @default.
- W4226168429 hasConceptScore W4226168429C13481523 @default.
- W4226168429 hasConceptScore W4226168429C154945302 @default.
- W4226168429 hasConceptScore W4226168429C162324750 @default.
- W4226168429 hasConceptScore W4226168429C177769412 @default.
- W4226168429 hasConceptScore W4226168429C187736073 @default.
- W4226168429 hasConceptScore W4226168429C2780451532 @default.
- W4226168429 hasConceptScore W4226168429C2781002164 @default.
- W4226168429 hasConceptScore W4226168429C28855332 @default.
- W4226168429 hasConceptScore W4226168429C41008148 @default.
- W4226168429 hasConceptScore W4226168429C51167844 @default.
- W4226168429 hasConceptScore W4226168429C9417928 @default.
- W4226168429 hasConceptScore W4226168429C94835093 @default.
- W4226168429 hasLocation W42261684291 @default.
- W4226168429 hasLocation W42261684292 @default.
- W4226168429 hasOpenAccess W4226168429 @default.
- W4226168429 hasPrimaryLocation W42261684291 @default.
- W4226168429 hasRelatedWork W2022517628 @default.
- W4226168429 hasRelatedWork W2374764536 @default.
- W4226168429 hasRelatedWork W2962928691 @default.
- W4226168429 hasRelatedWork W2964267629 @default.
- W4226168429 hasRelatedWork W2999283151 @default.
- W4226168429 hasRelatedWork W3199608561 @default.
- W4226168429 hasRelatedWork W4200632579 @default.
- W4226168429 hasRelatedWork W4221166828 @default.
- W4226168429 hasRelatedWork W4307321800 @default.
- W4226168429 hasRelatedWork W4319309271 @default.
- W4226168429 isParatext "false" @default.
- W4226168429 isRetracted "false" @default.
- W4226168429 workType "article" @default.