Ailia input shape(1, 3, 360, 640)
(Image from https://vision.middlebury.edu/stereo/data/scenes2003/)
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 crestereo.pyIf you want to specify the input image, put stereo images path after the --left, --right option, respectively.
$ python3 crestereo.py --left LEFT_IMAGE_PATH --right RIGTH_IMAGE_PATHIf you want to specify the input image, put the image path after the --left, --right option.
You can use --savepath option to change the name of the output file to save.
$ python3 crestereo.py --left LEFT_IMAGE_PATH --right RIGTH_IMAGE_PATH --savepath SAVE_IMAGE_PATHBy adding the --video_left, --video_rigth option, you can input the video.
If you pass 0 as an argument to RIGTH_VIDEO_PATH,1 as an argument to LEFT_VIDEO_PATH , you can use the webcam input instead of the video file.
$ python3 monodepth2.py --video_rigth RIGTH_VIDEO_PATH --video_left LEFT_VIDEO_PATHONNX-CREStereo-Depth-Estimation
PyTorch 1.11.0
ONNX opset = 12


