Logo image
CATTUS: A 4K-UHD 30FPS Deep Image Processor for Channel Attention Equipped U-Net Acceleration in 16nm FinFET
會議論文集

CATTUS: A 4K-UHD 30FPS Deep Image Processor for Channel Attention Equipped U-Net Acceleration in 16nm FinFET

Yong-Tai Chen, Yen-Ting Chiu, Yu-Chou Lee, Po-Yen Lu, Mu-Hsien Lee, Jia-Syuan Chen, Kuan-Hsien Ho, Hong-Chuan Liao, Chieh-Cheng Chen 和 Chao-Tsung Huang
Digest of technical papers - Symposium on VLSI Technology, 頁碼.1-3
08/06/2025

摘要

FinFETs Image reconstruction Imaging Memory management Mobile handsets Real-time systems Throughput Tunneling Very large scale integration Energy Efficiency
This paper presents CATTUS, the first deep image processor to integrate U-Net acceleration with channel attention, enabling high-quality image deblurring on mobile devices. It achieves throughput of 4K-UHD 30fps with 4.4 TOPS/W of energy efficiency. The processor realizes image reconstruction on challenging tasks such as deblurring with three key features: 1) pyramid layer fusion (PLF) for efficient U-net acceleration, 2) skip tunneling that minimizes external memory access (EMA) by optimizing skip connections, 3) lookahead channel attention (LCA) to reduce EMA for aggregating global information. Fabricated in 16nm FinFET technology, CATTUS delivers a 1.53× throughput improvement and 82% EMA reduction compared to baseline implementation, demonstrating real-time imaging across multiple applications.

相關連結

指標

1 檢視次數

詳細資料

Logo image