road-segmentation-adas

road-segmentation-adas-0001

Input

(Image from https://www.pexels.com/ja-jp/video/854669/)

Shape : (1, 512, 896, 3) BGR channel order

Output

Shape : (1, 512, 896, 4)

Category

CATEGORY = {
    'BG': 0,
    'road': 1,
    'curb': 2,
    'mark': 3,
}

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 road-segmentation-adas.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 road-segmentation-adas.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 road-segmentation-adas.py --video VIDEO_PATH

Reference

• OpenVINO - Open Model Zoo repository - road-segmentation-adas-0001
• OpenVINO - road-segmentation-adas-0001

Framework

OpenVINO

Model Format

ONNX opset=11

Netron

road-segmentation-adas-0001.onnx.prototxt