Image Super-Resolution via Dual-State Recurrent Networks
arXiv:1805.02704
Abstract
Advances in image super-resolution (SR) have recently benefited significantly from rapid developments in deep neural networks. Inspired by these recent discoveries, we note that many state-of-the-art deep SR architectures can be reformulated as a single-state recurrent neural network (RNN) with finite unfoldings. In this paper, we explore new structures for SR based on this compact RNN view, leading us to a dual-state design, the Dual-State Recurrent Network (DSRN). Compared to its single state counterparts that operate at a fixed spatial resolution, DSRN exploits both low-resolution (LR) and high-resolution (HR) signals jointly. Recurrent signals are exchanged between these states in both directions (both LR to HR and HR to LR) via delayed feedback. Extensive quantitative and qualitative evaluations on benchmark datasets and on a recent challenge demonstrate that the proposed DSRN performs favorably against state-of-the-art algorithms in terms of both memory consumption and predictive accuracy.
References in corpus (13)
- Deep Residual Learning for Image Recognition
- Densely Connected Convolutional Networks
- Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections
- Residual Networks Behave Like Ensembles of Relatively Shallow Networks
- Deep Networks with Stochastic Depth
- Bridging the Gaps Between Residual Learning, Recurrent Neural Networks and Visual Cortex
- Deep Networks for Image Super-Resolution with Sparse Prior
- Architectural Complexity Measures of Recurrent Neural Networks
- Accurate Image Super-Resolution Using Very Deep Convolutional Networks
- Dual Path Networks
- Sharing Residual Units Through Collective Tensor Factorization in Deep Neural Networks
- Self-Tuned Deep Super Resolution
- Beyond Deep Residual Learning for Image Restoration: Persistent Homology-Guided Manifold Simplification