super-resolution 0.2.3

Library generating 4x/8x super resoltion using deep convolutional neural networks.

Super Resolution

Image Super-Resolution using Deep Convolutional Neural Networks.

Installing

Install and update using pip:

pip3 install super-resolution

Or

git checkout https://github.com/fengwang/super_resolution.git
cd super_resolution
python3 -m pip install -e .

Usage

Command line:

super_resolution INPUT_IMAGE_PATH OUTPUT_IMAGE_PATH_4X

Using Python API:

# uncomment the follow three lines if you have a Nvidia GPU but you do not want to enable it.
#import os
#os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
#os.environ["CUDA_VISIBLE_DEVICES"]=''

from super_resolution import cartoon_upsampling_4x
large_image = cartoon_upsampling_4x( './a_tiny_input_image.png', './a_4x_larger_output_image.png' )

from super_resolution import cartoon_upsampling_8x
large_image = cartoon_upsampling_8x( './a_tiny_input_image.png', './a_8x_larger_output_image.png' )

Details

• The super resolution model is inherited from Ledig C, Theis L, Huszár F, et al. Photo-realistic single image super-resolution using a generative adversarial network, Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4681-4690.
• The training images are downloaded from Konachan (NSFW).

License

• BSD