TY - GEN
T1 - SMM-Conv
T2 - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2022
AU - Ofir, Amir
AU - Ben-Artzi, Gil
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022/6/22
Y1 - 2022/6/22
N2 - We present a novel approach for accelerating convolutions during inference for CPU-based architectures. The most common method of computation involves packing the image into the columns of a matrix (im2col) and performing general matrix multiplication (GEMM) with a matrix of weights. This results in two main drawbacks: (a) im2col requires a large memory buffer and can experience inefficient memory access, and (b) while GEMM is highly optimized for scientific matrices multiplications, it is not well suited for convolutions. We propose an approach that takes advantage of scalar-matrix multiplication and reduces memory overhead. Our experiments with commonly used network architectures demonstrate a significant speedup compared to existing indirect methods.
AB - We present a novel approach for accelerating convolutions during inference for CPU-based architectures. The most common method of computation involves packing the image into the columns of a matrix (im2col) and performing general matrix multiplication (GEMM) with a matrix of weights. This results in two main drawbacks: (a) im2col requires a large memory buffer and can experience inefficient memory access, and (b) while GEMM is highly optimized for scientific matrices multiplications, it is not well suited for convolutions. We propose an approach that takes advantage of scalar-matrix multiplication and reduces memory overhead. Our experiments with commonly used network architectures demonstrate a significant speedup compared to existing indirect methods.
UR - http://www.scopus.com/inward/record.url?scp=85137793500&partnerID=8YFLogxK
U2 - 10.1109/CVPRW56347.2022.00346
DO - 10.1109/CVPRW56347.2022.00346
M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???
AN - SCOPUS:85137793500
T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
SP - 3066
EP - 3074
BT - Proceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2022
PB - IEEE Computer Society
Y2 - 19 June 2022 through 20 June 2022
ER -