Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312596472> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4312596472 abstract "In this paper, we present an advanced algorithm-hardware co-optimization method for designing an efficient accelerator architecture for image signal processing (ISP) with deep neural networks (DNNs). Based on the systolic-array structure, for performing the target network model, we newly introduce two evaluation metrics, each of which is dedicated to fairly representing either the processing speed or the energy consumption. Then, the overall evaluation metric is defined to test each systolic array, finding the initial array configuration for the given number of total multipliers. From the initial array, several array-scaling methods are then presented to find the most cost-efficient array structure. In addition, the original ML model is adjusted to further enhance the overall efficiency with subtle quality drops of image outputs. Implementation results in 28nm CMOS technology show that the proposed co-optimization method successfully finds the cost-efficient systolic accelerator architecture for ISP applications, improving the energy efficiency by 51% compared to the straightforward array design." @default.
- W4312596472 created "2023-01-05" @default.
- W4312596472 creator A5029202746 @default.
- W4312596472 creator A5030748858 @default.
- W4312596472 creator A5082220016 @default.
- W4312596472 date "2022-05-28" @default.
- W4312596472 modified "2023-09-27" @default.
- W4312596472 title "Algorithm-Hardware Co-Optimization for Cost-Efficient ML-based ISP Accelerator" @default.
- W4312596472 cites W2133665775 @default.
- W4312596472 cites W2604319603 @default.
- W4312596472 cites W2793628756 @default.
- W4312596472 cites W2801748224 @default.
- W4312596472 cites W2945146780 @default.
- W4312596472 cites W2952046917 @default.
- W4312596472 cites W2989199275 @default.
- W4312596472 cites W3015208590 @default.
- W4312596472 cites W3024621361 @default.
- W4312596472 cites W3035381835 @default.
- W4312596472 cites W3047390932 @default.
- W4312596472 cites W3097528158 @default.
- W4312596472 cites W3100498948 @default.
- W4312596472 cites W3121915008 @default.
- W4312596472 cites W3134235012 @default.
- W4312596472 cites W3135799625 @default.
- W4312596472 cites W3175157054 @default.
- W4312596472 cites W3177327133 @default.
- W4312596472 doi "https://doi.org/10.1109/iscas48785.2022.9937743" @default.
- W4312596472 hasPublicationYear "2022" @default.
- W4312596472 type Work @default.
- W4312596472 citedByCount "0" @default.
- W4312596472 crossrefType "proceedings-article" @default.
- W4312596472 hasAuthorship W4312596472A5029202746 @default.
- W4312596472 hasAuthorship W4312596472A5030748858 @default.
- W4312596472 hasAuthorship W4312596472A5082220016 @default.
- W4312596472 hasConcept C113775141 @default.
- W4312596472 hasConcept C11413529 @default.
- W4312596472 hasConcept C119599485 @default.
- W4312596472 hasConcept C127413603 @default.
- W4312596472 hasConcept C13164978 @default.
- W4312596472 hasConcept C14580979 @default.
- W4312596472 hasConcept C149635348 @default.
- W4312596472 hasConcept C150741067 @default.
- W4312596472 hasConcept C176217482 @default.
- W4312596472 hasConcept C21547014 @default.
- W4312596472 hasConcept C24326235 @default.
- W4312596472 hasConcept C2742236 @default.
- W4312596472 hasConcept C41008148 @default.
- W4312596472 hasConcept C42935608 @default.
- W4312596472 hasConcept C46362747 @default.
- W4312596472 hasConcept C9390403 @default.
- W4312596472 hasConceptScore W4312596472C113775141 @default.
- W4312596472 hasConceptScore W4312596472C11413529 @default.
- W4312596472 hasConceptScore W4312596472C119599485 @default.
- W4312596472 hasConceptScore W4312596472C127413603 @default.
- W4312596472 hasConceptScore W4312596472C13164978 @default.
- W4312596472 hasConceptScore W4312596472C14580979 @default.
- W4312596472 hasConceptScore W4312596472C149635348 @default.
- W4312596472 hasConceptScore W4312596472C150741067 @default.
- W4312596472 hasConceptScore W4312596472C176217482 @default.
- W4312596472 hasConceptScore W4312596472C21547014 @default.
- W4312596472 hasConceptScore W4312596472C24326235 @default.
- W4312596472 hasConceptScore W4312596472C2742236 @default.
- W4312596472 hasConceptScore W4312596472C41008148 @default.
- W4312596472 hasConceptScore W4312596472C42935608 @default.
- W4312596472 hasConceptScore W4312596472C46362747 @default.
- W4312596472 hasConceptScore W4312596472C9390403 @default.
- W4312596472 hasFunder F4320322120 @default.
- W4312596472 hasLocation W43125964721 @default.
- W4312596472 hasOpenAccess W4312596472 @default.
- W4312596472 hasPrimaryLocation W43125964721 @default.
- W4312596472 hasRelatedWork W2365743651 @default.
- W4312596472 hasRelatedWork W2908855967 @default.
- W4312596472 hasRelatedWork W3101029515 @default.
- W4312596472 hasRelatedWork W3129147932 @default.
- W4312596472 hasRelatedWork W3180138673 @default.
- W4312596472 hasRelatedWork W4220667423 @default.
- W4312596472 hasRelatedWork W4226443758 @default.
- W4312596472 hasRelatedWork W4285346803 @default.
- W4312596472 hasRelatedWork W4306645249 @default.
- W4312596472 hasRelatedWork W4308067493 @default.
- W4312596472 isParatext "false" @default.
- W4312596472 isRetracted "false" @default.
- W4312596472 workType "article" @default.