View in EDS

High Dimensional Convolution Acceleration via Tensor Decomposition.

Saved in:
Bibliographic Details
Title: High Dimensional Convolution Acceleration via Tensor Decomposition.
Authors: Du, Xinyu1 (AUTHOR) jinhuitang@njust.edu.cn, Gu, Zichen2 (AUTHOR) guzichen2018@rails.cn, Dai, Longquan3 (AUTHOR) dailongquan@njust.edu.cn, Tang, Jinhui3 (AUTHOR) jinhuitang@njust.edu.cn
Source: Journal of Circuits, Systems & Computers. 2022, Vol. 31 Issue 10, p1-21. 21p.
Subjects: Computer vision, Computational complexity, Problem solving
Abstract: The high-dimensional convolution, in either linear or nonlinear form, has been employed in a wide range of computer vision solutions due to its beneficial smoothing property. However, its full-kernel implementation is extremely slow. We do need a fast algorithm for this important operation. To solve this problem, we propose an acceleration pipeline assembled by three steps: d -D nonlinear convolution ⇒ 1 d -D linear convolution ⇒ 2 1-D dimensional convolution ⇒ 3 1-D recursive box filter. Thanks to the low computational complexity of box filtering, we speed up the computation significantly. Roughly speaking, our contribution is two-fold: (1) establishing the connection between the high-dimensional convolution acceleration algorithm and tensor decomposition; (2) propose total four acceleration technologies including demultiplexing–blurring–multiplexing framework, convolution decomposition, periodic tensorization and recursively box filtering to compose our acceleration pipeline under the line of the above connection. The effectiveness of these techniques is demonstrated in various comparisons and experiments. The running times of various applications are largely shortened from several minutes to fewer seconds or less. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Circuits, Systems & Computers is the property of World Scientific Publishing Company and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:The high-dimensional convolution, in either linear or nonlinear form, has been employed in a wide range of computer vision solutions due to its beneficial smoothing property. However, its full-kernel implementation is extremely slow. We do need a fast algorithm for this important operation. To solve this problem, we propose an acceleration pipeline assembled by three steps: d -D nonlinear convolution ⇒ 1 d -D linear convolution ⇒ 2 1-D dimensional convolution ⇒ 3 1-D recursive box filter. Thanks to the low computational complexity of box filtering, we speed up the computation significantly. Roughly speaking, our contribution is two-fold: (1) establishing the connection between the high-dimensional convolution acceleration algorithm and tensor decomposition; (2) propose total four acceleration technologies including demultiplexing–blurring–multiplexing framework, convolution decomposition, periodic tensorization and recursively box filtering to compose our acceleration pipeline under the line of the above connection. The effectiveness of these techniques is demonstrated in various comparisons and experiments. The running times of various applications are largely shortened from several minutes to fewer seconds or less. [ABSTRACT FROM AUTHOR]
ISSN:02181266
DOI:10.1142/S0218126622501870