Skip to content

Latest commit

 

History

History

rcan-it

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 
 
 

RCAN-it

Input

Input

(Image from https://github.com/sanghyun-son/EDSR-PyTorch/blob/master/test/0853x4.png)

Ailia input shape : (1, 3, IMAGE_HEIGHT, IMAGE_WIDTH)

Output

Output

Ailia output shape : (1, 3, IMAGE_HEIGHT * scale, IMAGE_WIDTH * scale)

default : scale=2

Usage

Automatically downloads the onnx and prototxt files when running. It is necessary to be connected to the Internet while downloading.

For the sample image with twice the resolution,

$ python3 rcan-it.py

If you want to specify the input image, put the image path after the --input option.
You can use --savepath option to change the name of the output file to save.

$ python3 rcan-it.py --input IMAGE_PATH --savepath SAVE_IMAGE_PATH

If you want to specify the scale for the resolution, put the scale after the --scale option.
Choose the scale in [2, 3, 4].

$ python3 rcan-it.py --scale SCALE 

By adding the --video option, you can input the video.
If you pass 0 as an argument to VIDEO_PATH, you can use the webcam input instead of the video file.

$ python3 rcan-it.py --video VIDEO_PATH

Reference

Revisiting RCAN: Improved Training for Image Super-Resolution

Framework

Pytorch 1.11.0

Model Format

ONNX opset = 12

Netron

rcan-it_scale2.onnx.prototxt

rcan-it_scale3.onnx.prototxt

rcan-it_scale4.onnx.prototxt