Neural Architecture Search for Image Super-Resolution Using Densely Constructed Search Space: DeCoNAS

TL;DR

This paper introduces DeCoNAS, a neural architecture search method for image super-resolution that automatically finds lightweight, densely connected networks outperforming existing handcrafted and NAS-designed models.

Contribution

It expands NAS to super-resolution, proposing a hierarchical search strategy with a complexity-based penalty for multi-objective optimization.

Findings

DeCoNASNet outperforms state-of-the-art lightweight super-resolution networks.

The hierarchical search effectively balances local and global feature connections.

The complexity-based penalty improves the search for efficient architectures.

Abstract

The recent progress of deep convolutional neural networks has enabled great success in single image super-resolution (SISR) and many other vision tasks. Their performances are also being increased by deepening the networks and developing more sophisticated network structures. However, finding an optimal structure for the given problem is a difficult task, even for human experts. For this reason, neural architecture search (NAS) methods have been introduced, which automate the procedure of constructing the structures. In this paper, we expand the NAS to the super-resolution domain and find a lightweight densely connected network named DeCoNASNet. We use a hierarchical search strategy to find the best connection with local and global features. In this process, we define a complexity-based penalty for solving image super-resolution, which can be considered a multi-objective problem. Experiments show that our DeCoNASNet outperforms the state-of-the-art lightweight super-resolution networks designed by handcraft methods and existing NAS-based design.