Fine-grained neural architecture search for image super-resolution
- Authors
- Kim, Heewon; Hong, Seokil; Han, Bohyung; Myeong, Heesoo; Lee, Kyoung Mu
- Issue Date
- Nov-2022
- Publisher
- ACADEMIC PRESS INC ELSEVIER SCIENCE
- Keywords
- Image super-resolution; Neural architecture search; Convolutional neural network
- Citation
- JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, v.89
- Journal Title
- JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION
- Volume
- 89
- URI
- http://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/43434
- DOI
- 10.1016/j.jvcir.2022.103654
- ISSN
- 1047-3203
- Abstract
- Designing efficient deep neural networks has achieved great interest in image super-resolution (SR). However, exploring diverse network structures is computationally expensive. More importantly, each layer in a network has a distinct role that leads to the design of a specialized structure. In this work, we present a novel neural architecture search (NAS) algorithm that efficiently explores layer-wise structures. Specifically, we construct a supernet allowing flexibility in choosing the number of channels and per-channel activation functions according to the role of each layer. The search process runs efficiently via channel pruning since gradient descent jointly optimizes the Mult-Adds and the accuracy of the searched models. We facilitate estimating the model Mult-Adds in a differentiable manner using relaxations in the backward pass. The searched model, named FGNAS, outperforms the state-of-the-art NAS-based SR methods by a large margin.
- Files in This Item
-
Go to Link
- Appears in
Collections - College of Information Technology > Global School of Media > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.