Skip to content

Latest commit

 

History

History
 
 

fast-depth

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
data/img
data/img
 
 
img
img
 
 
transforms
transforms
 
 
LICENSE
LICENSE
 
 
README.md
README.md
 
 
fast-depth.py
fast-depth.py
 
 
fast_depth_utils.py
fast_depth_utils.py
 
 
output.png
output.png
 
 

README.md

FastDepth: Fast Monocular Depth Estimation on Embedded Systems

input

  • input image (1x3x250x250)

(Extracted from NYU Depth V2 dataset in HDF5 format.)

output

  • output image (1x1x224x224)

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 fast-depth.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 directory of the output file to be saved.

$ python3 fast-depth.py --input IMAGE_PATH --savepath SAVE_IMAGE_PATH

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

$ python3 fast-depth.py --video VIDEO_PATH

Reference

ICRA 2019 "FastDepth: Fast Monocular Depth Estimation on Embedded Systems"

Framework

PyTorch

Model Format

ONNX opset = 11

Netron