We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Hard parameter sharing for compressing dense-connection-based image restoration network.
- Authors
Tian, Xiang; Zheng, Bolun; Li, Shengyu; Yan, Chenggang; Zhang, Jiyong; Sun, Yaoqi; Shen, Tao; Xiao, Mang
- Abstract
The dense connection is a powerful technique to build wider and deeper convolution neural networks (CNNs) for handling several computer vision tasks. Despite the excellent performance, it consumes numerous parameters and produces a large weight model file. We studied the distribution of convolution layers and proposed a hard parameter sharing approach known as convolution pool (CP) for compressing dense-connection-based image restoration CNN models. CP is used to reallocate the parameters to specific convolution layers to ensure that some can be shared in different layers. We design a set of dense-connection-based baselines for three typical image restoration tasks, including image denoising, super-resolution, and JPEG deblocking, to validate the performance of the proposed method. Moreover, we comprehensively analyze the potential problems by introducing CP, including group convolution, dilated convolution, and modeling efficiency. Experimental results demonstrate that the proposed method can efficiently achieve an impressive compression rate with negligible performance reduction.
- Subjects
IMAGE reconstruction; CONVOLUTIONAL neural networks; IMAGE denoising; COMPUTER vision
- Publication
Journal of Electronic Imaging, 2021, Vol 30, Issue 5, p53025
- ISSN
1017-9909
- Publication type
Article
- DOI
10.1117/1.JEI.30.5.053025