bobliu20/superresolution_caffe

Implementate super resolution in deep learning

Super Resolution

All of this methods is implementation in caffe python.

Instruction

• VDSR (Very Deep network for Super-Resolution) is an end-to-end network with 20 convolutional layers for single image super-resolution. [Web]

• SRCNN (Image Super-Resolution Using Deep Convolutional Networks) [Web]

• FSRCNN (Accelerating the Super-Resolution Convolutional Neural Network) [Web]

• MY_SR was modified from VDSR and SRCNN. Better performance for this task.

How to training in my code

First of all. I use Super-Resolution to do data matrix reader (Enhance picture quality). My platform is Ubuntu 14.04 with pycaffe.

1. Install libdmtx

Download source code from this site. And then following readme to build and install libdmtx. Also do python setup.py install to install python wrapper name pydmtx.

2. Set environment variable

    export LD_LIBRARY_PATH=/usr/local/lib/:$LD_LIBRARY_PATH
    export PYTHONPATH=`pwd`:$PYTHONPATH

3. Start to training

    caffe train --solver=solver.prototxt --gpu=0

Result compare

• VDSR

VS

• SRCNN

VS

• MY_SR

VS