· 2016
Accurate Image Super-Resolution Using Very Deep Convolutional Networks
- 7,634citations
- 2016year
Short summary
A 20-layer deep convolutional network, inspired by VGG-net, achieves state-of-the-art single-image super-resolution (SR) accuracy by cascading small filters to exploit large-scale contextual information.
AI-generated from the title and abstract; the full text is not read.
Key points
- A 20-layer deep convolutional network is proposed for single-image super-resolution (SR).
- Increasing network depth significantly improves SR accuracy.
- Cascading small filters in a deep network efficiently exploits large-scale contextual information.
- A new training method uses residual learning with extremely high learning rates (10^4x higher than SRCNN) and gradient clipping.
AI-generated from the title and abstract; the full text is not read.
Abstract
We present a highly accurate single-image superresolution (SR) method. Our method uses a very deep convolutional network inspired by VGG-net used for ImageNet classification [19]. We find increasing our network depth shows a significant improvement in accuracy. Our final model uses 20 weight layers. By cascading small filters many times in a deep network structure, contextual information over large image regions is exploited in an efficient way. With very deep networks, however, convergence speed becomes a critical issue during training. We propose a simple yet effective training procedure. We learn residuals only and use extremely high learning rates (104 times higher than SRCNN [6]) enabled by adjustable gradient clipping. Our proposed method performs better than existing methods in accuracy and visual improvements in our results are easily noticeable.
The authors' abstract, as published at the source., 2016 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionComputer Science