yet-another-anime-segmenter

Yet-Another-Anime-Segmenter

Input

(Image from https://unity-chan.com/download/index.php and licensed under © Unity Technologies Japan/UCL)

• ailia input shape: (1, 3, height, width) RGB channel order
  • Accepts any raw image height and width but needs to be resized and padded accordingly (see preprocessing method)
• Pixel value range: Normalized values (see preprocessing method)

Output

• ailia Predict API output: Raw classification, kernel and mask predictions
• With helper functions, classification scores, predicted classes, predicted masks, predicted bounding boxes can be obtained.

Usage

Automatically downloads the onnx and prototxt files on the first run. It is necessary to be connected to the Internet while downloading.

For the sample image,

$ python3 yet-another-anime-segmenter.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 yet-another-anime-segmenter.py --input IMAGE_PATH --savepath SAVE_IMAGE_PATH

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 yet-another-anime-segmenter.py --video VIDEO_PATH --savepath SAVE_VIDEO_PATH

To use the ONNX Runtime instead of ailia.

$ python3 yet-another-anime-segmenter.py --onnx

Reference

• Yet-Another-Anime-Segmenter
• AdelaiDet
• detectron2

Framework

PyTorch 1.7.1

Model Format

ONNX opset = 11

Netron

yaas_solov2.onnx.prototxt