Software installation

We suggest to use Ubuntu 22.04 for the software stack. Since VisionPilot requires AI packages, we need to install the following packages first.

Prerequisites

• CUDA: Optional for GPU processing.
• OpenCV: For image and video processing.
  • Ubuntu: sudo apt install libopencv-dev
• ONNX Runtime: For model inference.
  • Download from the GitHub release
• LibTorch: For tensor manipulation capabilities.
  • Download from the PyTorch website

Models

The following models will be used in the pipeline.

• SceneSeg - segmentation of foreground objects:
  • Link to Download Pytorch Model Weights *.pth
  • Link to Download Traced Pytorch Model *.pt
  • Link to Download ONNX FP32 Weights *.onnx
• Scene3D - monocular depth estimation
  • Link to Download Pytorch Model Weights *.pth
  • Link to Download Traced Pytorch Model *.pt
  • Link to Download ONNX FP32 Weights *.onnx
• DomainSeg - segmentation of construction objects
  • Link to Download Pytorch Model Weights *.pth
  • Link to Download Traced Pytorch Model *.pt
  • Link to Download ONNX FP32 Weights *.onnx