https://github.com/edubart/superesnet
Single image super resolution in Torch
https://github.com/edubart/superesnet
super-resolution torch7 vdsr
Last synced: about 2 months ago
JSON representation
Single image super resolution in Torch
- Host: GitHub
- URL: https://github.com/edubart/superesnet
- Owner: edubart
- Created: 2017-05-09T16:05:45.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2017-05-09T16:09:28.000Z (about 8 years ago)
- Last Synced: 2025-02-10T02:42:12.906Z (4 months ago)
- Topics: super-resolution, torch7, vdsr
- Language: Lua
- Size: 14 MB
- Stars: 0
- Watchers: 4
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Experiments with super resolution.
Currently modified versions of VDSR and LapSRN for 2x are implemented,
trained model on game sprites is on models folders, train parameters
are in experiments folder.
Papers based on:
- [Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution](https://arxiv.org/abs/1704.03915)
- [Accurate Image Super-Resolution Using Very Deep Convolutional Networks](https://arxiv.org/abs/1511.04587)
- [Online Learning Rate Adaptation with Hypergradient Descent](https://arxiv.org/abs/1703.04782)
- [Stochastic Gradient Descent with Warm Restarts](https://arxiv.org/abs/1608.03983)
Other references:
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://arxiv.org/abs/1502.03167)
- [Deconvolution and Checkerboard Artifacts](http://distill.pub/2016/deconv-checkerboard/)