04 Hybrid Vision algo

Main Features

  • 🚀 High Performance: Optimized C++17 implementation, supports real-time event stream processing
  • 🧠 Multiple Denoising Algorithms: Includes traditional and deep learning methods
  • 🔧 Easy Integration: Provides CMake configuration and pkg-config support
  • 📦 Modular Design: Clear module division, easy to extend and maintain
  • 🎯 Practical Examples: Rich example code and documentation

Algorithm Modules

Denoising Algorithms (Denoise)

  1. Double Window Filter

    • Uses two circular buffers to classify events
    • Suitable for real-time processing in low noise environments

  2. Event Flow Filter

    • Noise suppression based on event flow density and flow velocity characteristics
    • Suitable for denoising in motion scenes

  3. Khodamoradi Denoiser

    • Classical denoising algorithm based on spatiotemporal neighborhood
    • General solution balancing performance and effect

  4. Reclusive Event Denoiser

    • Recursive event processing algorithm
    • Suitable for complex noise environments

  5. Time Surface Denoiser

    • Denoising method based on time surface
    • Maintains event timing characteristics

  6. Yang Noise Filter

    • Noise filtering algorithm proposed by Yang et al.
    • Efficient real-time processing capability

  7. MLP Filter (Optional)

    • Intelligent denoising based on deep learning
    • Requires PyTorch support
    • Suitable for high-precision denoising in complex scenes

Computer Vision (CV)

  • Basic image processing algorithms
  • Feature extraction and matching

3D Vision (CV3D)

  • Stereo vision algorithms
  • Depth estimation
  • 3D reconstruction

Image Restoration (Restoration)

  • Image enhancement algorithms
  • Deblurring and super-resolution

System Requirements

Essential Dependencies

  • Ubuntu == 22.04
  • CMake >= 3.16
  • C++17 compatible compiler
  • Openeb SDK
  • Eigen3

Optional Dependencies

  • PyTorch (For MLP filter)
  • CUDA (GPU acceleration support)

Installation Guide

1. Install Dependencies

Ubuntu/Debian

# Install basic dependencies
sudo apt update
sudo apt install cmake build-essential libeigen3-dev

# Install HV Toolkits
# Pull submodule
git submodule update --init
# Enter submodule directory
cd external/shimetapi_hybrid_vision_toolkit/
sudo ./install_libs.sh

# Install Openeb
curl -L https://propheseeai.jfrog.io/artifactory/api/security/keypair/prophesee-gpg/public >/tmp/propheseeai.jfrog.op.asc
sudo cp /tmp/propheseeai.jfrog.op.asc /etc/apt/trusted.gpg.d
sudo add-apt-repository 'https://propheseeai.jfrog.io/artifactory/openeb-debian/'
sudo apt update
sudo apt -y install metavision-openeb