Revisiting Temporal Modeling for Video Super-resolution
Preprint
Abstract
Video super-resolution plays an important role in surveillance video analysis and ultra-high-definition video display, which has drawn much attention in both the research and industrial communities. Although many deep learning-based VSR methods have been proposed, it is hard to directly compare these methods since the different loss functions and training datasets have a significant impact on the super-resolution results. In this work, we carefully study and compare three temporal modeling methods (2D CNN with early fusion, 3D CNN with slow fusion and Recurrent Neural Network) for video super-resolution. We also propose a novel Recurrent Residual Network (RRN) for efficient video super-resolution, where residual learning is utilized to stabilize the training of RNN and meanwhile to boost the super-resolution performance. Extensive experiments show that the proposed RRN is highly computational efficiency and produces temporal consistent VSR results with finer details than other temporal modeling methods. Besides, the proposed method achieves state-of-the-art results on several widely used benchmarks.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- Video Super-Resolution with Recurrent Structure-Detail Network2020
- Video Enhancement with Task-Oriented FlowInternational Journal of Computer Vision, 2019
- Image Super-Resolution Using Very Deep Residual Channel Attention Networks2018
- Residual Dense Network for Image Super-Resolution2018
- Deep Back-Projection Networks for Super-Resolution2018
- Real-Time Video Super-Resolution with Spatio-Temporal Networks and Motion Compensation2017
- Image Super-Resolution via Deep Recursive Residual Network2017
- Deeply-Recursive Convolutional Network for Image Super-Resolution2016
- Accurate Image Super-Resolution Using Very Deep Convolutional Networks2016
- Learning a Deep Convolutional Network for Image Super-Resolution2014