Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288092016> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W4288092016 abstract "With the growth of computer vision based applications and services, an explosive amount of images have been uploaded to cloud servers which host such computer vision algorithms, usually in the form of deep learning models. JPEG has been used as the {em de facto} compression and encapsulation method before one uploads the images, due to its wide adaptation. However, standard JPEG configuration does not always perform well for compressing images that are to be processed by a deep learning model, e.g., the standard quality level of JPEG leads to 50% of size overhead (compared with the best quality level selection) on ImageNet under the same inference accuracy in popular computer vision models including InceptionNet, ResNet, etc. Knowing this, designing a better JPEG configuration for online computer vision services is still extremely challenging: 1) Cloud-based computer vision models are usually a black box to end-users; thus it is difficult to design JPEG configuration without knowing their model structures. 2) JPEG configuration has to change when different users use it. In this paper, we propose a reinforcement learning based JPEG configuration framework. In particular, we design an agent that adaptively chooses the compression level according to the input image's features and backend deep learning models. Then we train the agent in a reinforcement learning way to adapt it for different deep learning cloud services that act as the {em interactive training environment} and feeding a reward with comprehensive consideration of accuracy and data size. In our real-world evaluation on Amazon Rekognition, Face++ and Baidu Vision, our approach can reduce the size of images by 1/2 -- 1/3 while the overall classification accuracy only decreases slightly." @default.
- W4288092016 created "2022-07-28" @default.
- W4288092016 creator A5014011362 @default.
- W4288092016 creator A5034104790 @default.
- W4288092016 creator A5034630340 @default.
- W4288092016 creator A5036016529 @default.
- W4288092016 creator A5089553256 @default.
- W4288092016 date "2019-10-15" @default.
- W4288092016 modified "2023-09-28" @default.
- W4288092016 title "AdaCompress" @default.
- W4288092016 cites W1605005685 @default.
- W4288092016 cites W2013305145 @default.
- W4288092016 cites W2117539524 @default.
- W4288092016 cites W2276892413 @default.
- W4288092016 cites W2468875367 @default.
- W4288092016 cites W2513158035 @default.
- W4288092016 cites W2568772110 @default.
- W4288092016 cites W2577678996 @default.
- W4288092016 cites W2591924527 @default.
- W4288092016 cites W2626129225 @default.
- W4288092016 cites W2792447253 @default.
- W4288092016 cites W2806506455 @default.
- W4288092016 cites W2962858109 @default.
- W4288092016 cites W2963149687 @default.
- W4288092016 cites W2963163009 @default.
- W4288092016 cites W2995127918 @default.
- W4288092016 cites W4235435541 @default.
- W4288092016 doi "https://doi.org/10.1145/3343031.3350874" @default.
- W4288092016 hasPublicationYear "2019" @default.
- W4288092016 type Work @default.
- W4288092016 citedByCount "5" @default.
- W4288092016 countsByYear W42880920162021 @default.
- W4288092016 countsByYear W42880920162022 @default.
- W4288092016 countsByYear W42880920162023 @default.
- W4288092016 crossrefType "proceedings-article" @default.
- W4288092016 hasAuthorship W4288092016A5014011362 @default.
- W4288092016 hasAuthorship W4288092016A5034104790 @default.
- W4288092016 hasAuthorship W4288092016A5034630340 @default.
- W4288092016 hasAuthorship W4288092016A5036016529 @default.
- W4288092016 hasAuthorship W4288092016A5089553256 @default.
- W4288092016 hasBestOaLocation W42880920162 @default.
- W4288092016 hasConcept C108583219 @default.
- W4288092016 hasConcept C111919701 @default.
- W4288092016 hasConcept C119857082 @default.
- W4288092016 hasConcept C154945302 @default.
- W4288092016 hasConcept C198751489 @default.
- W4288092016 hasConcept C31258907 @default.
- W4288092016 hasConcept C31972630 @default.
- W4288092016 hasConcept C41008148 @default.
- W4288092016 hasConcept C71901391 @default.
- W4288092016 hasConcept C78548338 @default.
- W4288092016 hasConcept C79974875 @default.
- W4288092016 hasConcept C93996380 @default.
- W4288092016 hasConcept C97541855 @default.
- W4288092016 hasConceptScore W4288092016C108583219 @default.
- W4288092016 hasConceptScore W4288092016C111919701 @default.
- W4288092016 hasConceptScore W4288092016C119857082 @default.
- W4288092016 hasConceptScore W4288092016C154945302 @default.
- W4288092016 hasConceptScore W4288092016C198751489 @default.
- W4288092016 hasConceptScore W4288092016C31258907 @default.
- W4288092016 hasConceptScore W4288092016C31972630 @default.
- W4288092016 hasConceptScore W4288092016C41008148 @default.
- W4288092016 hasConceptScore W4288092016C71901391 @default.
- W4288092016 hasConceptScore W4288092016C78548338 @default.
- W4288092016 hasConceptScore W4288092016C79974875 @default.
- W4288092016 hasConceptScore W4288092016C93996380 @default.
- W4288092016 hasConceptScore W4288092016C97541855 @default.
- W4288092016 hasLocation W42880920161 @default.
- W4288092016 hasLocation W42880920162 @default.
- W4288092016 hasLocation W42880920163 @default.
- W4288092016 hasOpenAccess W4288092016 @default.
- W4288092016 hasPrimaryLocation W42880920161 @default.
- W4288092016 hasRelatedWork W3014300295 @default.
- W4288092016 hasRelatedWork W3164822677 @default.
- W4288092016 hasRelatedWork W4223943233 @default.
- W4288092016 hasRelatedWork W4225161397 @default.
- W4288092016 hasRelatedWork W4250304930 @default.
- W4288092016 hasRelatedWork W4312200629 @default.
- W4288092016 hasRelatedWork W4360585206 @default.
- W4288092016 hasRelatedWork W4364306694 @default.
- W4288092016 hasRelatedWork W4380075502 @default.
- W4288092016 hasRelatedWork W4380086463 @default.
- W4288092016 isParatext "false" @default.
- W4288092016 isRetracted "false" @default.
- W4288092016 workType "article" @default.