Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312121933> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W4312121933 abstract "Convolutional neural network (CNN) accelerators are being widely used for their efficiency, but they require a large amount of memory, leading to the use of a slow and power consuming external memory. This paper exploits two schemes to reduce the required memory amount and ultimately to implement a CNN of reasonable performance only with on-chip memory of a practical device like a low-end FPGA. To reduce the memory amount of the intermediate data, a stream-based line-buffer architecture and a dataflow for the architecture are proposed instead of the conventional frame-based architecture, where the amount of the intermediate data memory is proportional to the square of the input image size. The architecture consists of layer-dedicated blocks operating in a pipelined way with the input and output streams. Each convolutional layer block has a line buffer storing just a few rows of input data. The sizes of the line buffers are proportional to the width of the input image, so the architecture requires less intermediate data storage than the conventional frame-based architecture, especially in the trend of getting larger input size in modern object detection CNNs. In addition to the reduced intermediate data storage, the weight memory is reduced by the accelerator-aware pruning. The experimental results show that a whole object detection CNN can be implemented even on a low-end FPGA without an external memory. Compared to previous accelerators with similar object detection accuracy, the proposed accelerator reaches much higher throughput even with less FPGA resources of LUTs, registers, and DSPs, showing much higher efficiency. The trained models and implemented bit files are available at https://github.com/HyeongjuKang/accelerator-aware-pruning and https://github.com/HyeongjuKang/aocstream." @default.
- W4312121933 created "2023-01-04" @default.
- W4312121933 creator A5050847500 @default.
- W4312121933 date "2022-12-21" @default.
- W4312121933 modified "2023-09-26" @default.
- W4312121933 title "AoCStream: All-on-Chip CNN Accelerator With Stream-Based Line-Buffer Architecture" @default.
- W4312121933 doi "https://doi.org/10.48550/arxiv.2212.11438" @default.
- W4312121933 hasPublicationYear "2022" @default.
- W4312121933 type Work @default.
- W4312121933 citedByCount "0" @default.
- W4312121933 crossrefType "posted-content" @default.
- W4312121933 hasAuthorship W4312121933A5050847500 @default.
- W4312121933 hasBestOaLocation W43121219331 @default.
- W4312121933 hasConcept C111919701 @default.
- W4312121933 hasConcept C126042441 @default.
- W4312121933 hasConcept C149635348 @default.
- W4312121933 hasConcept C154945302 @default.
- W4312121933 hasConcept C157764524 @default.
- W4312121933 hasConcept C173608175 @default.
- W4312121933 hasConcept C2524010 @default.
- W4312121933 hasConcept C2777210771 @default.
- W4312121933 hasConcept C2779602883 @default.
- W4312121933 hasConcept C33923547 @default.
- W4312121933 hasConcept C41008148 @default.
- W4312121933 hasConcept C42935608 @default.
- W4312121933 hasConcept C555944384 @default.
- W4312121933 hasConcept C76155785 @default.
- W4312121933 hasConcept C81363708 @default.
- W4312121933 hasConcept C9390403 @default.
- W4312121933 hasConcept C96324660 @default.
- W4312121933 hasConceptScore W4312121933C111919701 @default.
- W4312121933 hasConceptScore W4312121933C126042441 @default.
- W4312121933 hasConceptScore W4312121933C149635348 @default.
- W4312121933 hasConceptScore W4312121933C154945302 @default.
- W4312121933 hasConceptScore W4312121933C157764524 @default.
- W4312121933 hasConceptScore W4312121933C173608175 @default.
- W4312121933 hasConceptScore W4312121933C2524010 @default.
- W4312121933 hasConceptScore W4312121933C2777210771 @default.
- W4312121933 hasConceptScore W4312121933C2779602883 @default.
- W4312121933 hasConceptScore W4312121933C33923547 @default.
- W4312121933 hasConceptScore W4312121933C41008148 @default.
- W4312121933 hasConceptScore W4312121933C42935608 @default.
- W4312121933 hasConceptScore W4312121933C555944384 @default.
- W4312121933 hasConceptScore W4312121933C76155785 @default.
- W4312121933 hasConceptScore W4312121933C81363708 @default.
- W4312121933 hasConceptScore W4312121933C9390403 @default.
- W4312121933 hasConceptScore W4312121933C96324660 @default.
- W4312121933 hasLocation W43121219331 @default.
- W4312121933 hasOpenAccess W4312121933 @default.
- W4312121933 hasPrimaryLocation W43121219331 @default.
- W4312121933 hasRelatedWork W1556657664 @default.
- W4312121933 hasRelatedWork W1572523360 @default.
- W4312121933 hasRelatedWork W2047588290 @default.
- W4312121933 hasRelatedWork W2054126032 @default.
- W4312121933 hasRelatedWork W2100229967 @default.
- W4312121933 hasRelatedWork W2623205115 @default.
- W4312121933 hasRelatedWork W2968111836 @default.
- W4312121933 hasRelatedWork W3092708771 @default.
- W4312121933 hasRelatedWork W97816082 @default.
- W4312121933 hasRelatedWork W1979576862 @default.
- W4312121933 isParatext "false" @default.
- W4312121933 isRetracted "false" @default.
- W4312121933 workType "article" @default.