.png){kind=icon}
Return
A Tile-Based Feature Map Compression Method for Multi-Stride Winograd Algorithm
.png){kind=icon}
PRE
AI
DOI：10.1109/tcsii.2026.3693787.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this brief, a Tile-based Feature map Compression Method (TFCM) was proposed to accelerate both convolutions with filter (3,2) (i.e., filter size is 3 and stride is 2) and (3,1) by using Winograd. Furthermore, the feasibility of TFCM was also discussed based on some popular CNNs. In addition, a Winograd-based Acceleration Unit (WAU) was designed to implement TFCM. The evaluation results show that our proposed methods can achieve 45.93%~75.28% multiplication savings when applied to VGG16, MobileNetV1. Moreover, our accelerator can accomplish 1.71 TOPS while deploying VGG16, achieving a $1.27\times $ to $2\times $ improvement in DSP efficiency per MHz compared with the state-of-the-art accelerators.
Keywords:
CNNs
winograd
feature map compression
Journal
I
IF:
4.9
Papers:
61
Citations:
0
Organization
Cited Papers
.png){kind=icon}
ShareSave
ShareSaveShareSave
View more