Skip to content

ibaiGorordo/ONNX-Unidepth-Monocular-Metric-Depth-Estimation

Repository files navigation

ONNX-UniDepth Monocular Metric Depth Estimation

!ONNX-UniDepth Monocular Metric Depth Estimation

Requirements

  • Check the requirements.txt file.
  • For ONNX, if you have a NVIDIA GPU, then install the onnxruntime-gpu, otherwise use the onnxruntime library.

Installation

git clone https://github.com/ibaiGorordo/ONNX-Unidepth-Monocular-Metric-Depth-Estimation.git
cd ONNX-Unidepth-Monocular-Metric-Depth-Estimation
pip install -r requirements.txt

ONNX Runtime

For Nvidia GPU computers: pip install onnxruntime-gpu

Otherwise: pip install onnxruntime

ONNX model

  • Download the model from Hugging Face and save it in the models folder.
  • Otherwise use this Google Colab notebook to convert the model: Open In Colab

Original UniDepth model

The original UniDepth model can be found in this repository: UniDepth Repository

  • The License of the models is Attribution-NonCommercial 4.0 International: License

Examples

  • Image inference:
python image_depth_estimation.py
  • Webcam inference:
python webcam_depth_estimation.py
  • Video inference:
python video_depth_estimation.py
python video_depth_estimation.py

!Unidepth Rerun SDK visualization

Original video: https://www.youtube.com/watch?v=8jsXi_51B40

References:

About

Python scripts performing Metric Monocular Depth Estimation using the Unidepth model in ONNX.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages