fcrn-depthprediction

FCRN-DepthPrediction

Input

(Image from https://pixabay.com/photos/bedroom-cupboard-bed-room-sofa-1872196/)

Shape : (1, 228, 304, 3)

Output

Shape : (128, 160, 1)

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 fcrn-depthprediction.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 fcrn-depthprediction.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 fcrn-depthprediction.py --video VIDEO_PATH

Reference

Deeper Depth Prediction with Fully Convolutional Residual Networks

Framework

TensorFlow

Model Format

ONNX opset=11

Netron

ResNet50UpProj.onnx.prototxt