English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Vision transformers (ViTs) are emerging with significantly improved accuracyin computer vision tasks. However, their complex architecture and enormouscomputation/storage demand impose urgent needs for new hardware acceleratordesign methodology. This work proposes an FPGA-aware automatic ViT accelerationframework based on the proposed mixed-scheme quantization. To the best of ourknowledge, this is the first FPGA-based ViT acceleration framework exploringmodel quantization. Compared with state-of-the-art ViT quantization work(algorithmic approach only without hardware acceleration), our quantizationachieves 0.47% to 1.36% higher Top-1 accuracy under the same bit-width.Compared with the 32-bit floating-point baseline FPGA accelerator, ouraccelerator achieves around 5.6x improvement on the frame rate (i.e., 56.8 FPSvs. 10.0 FPS) with 0.71% accuracy drop on ImageNet dataset for DeiT-base.

Auto-ViT-Acc: An FPGA-Aware Automatic Acceleration Framework for Vision Transformer with Mixed-Scheme Quantization