This repository provides a tensor visualization tool to visualize 3D tensor fields with built-in functions and performs efficient 3D tensor voting using parallel computations with CUDA.

1. Overview
2. Demo
3. Dependencies
4. Installation
5. Usage
6. Contributing

The repository focuses on:

• Tools for generating, processing, and visualizing tensor fields and volumetric data.
• Parallelized tensor voting computations using CUDA for GPU acceleration.
• Efficient parallelized eigendecomposition of tensors for 3D data.
• Synthetic dataset generation for validation and testing.

This project is designed for research in large-scale tensor voting frameworks, particularly for 3D data analysis and segmentation. For the complete repository (including the 2D version, image-to-tensor library, etc.) visit the STIM LAB Repository.

This demo shows the functionality of the TensorView3D program. A tensor field generated from Serial Block-Face Scanning Electron Microscopy (SBF-SEM) data is loaded into the program, where various tools and visualization techniques are applied to the tensor field.

• Python 3.8+
• Required Python packages:
  • numpy
  • scikit-image
  • scipy
• C++ Libraries:
  • OpenGL
  • FreeGLUT
  • Boost Program Options
  • Eigen3
  • GLEW
  • ImGui
  • GLFW3
• CUDA Toolkit (for GPU-accelerated computations)

• Windows OS (project does not currently support macOS)
• CUDA Toolkit: Install from NVIDIA CUDA Toolkit.
• CMake: Version 3.29 or higher.
• vcpkg (for managing C++ dependencies).

Install vcpkg to manage the required C++ packages:

git clone https://github.com/microsoft/vcpkg.git
cd vcpkg
bootstrap-vcpkg.bat

Run the following commands to install required libraries:

vcpkg install glfw3 glm imgui[core,glfw-binding,opengl3-binding] glew boost-program-options eigen3

These commands install:

• glfw3 for window handling
• glm for math operations
• imgui for the GUI
• GLEW for OpenGL extensions
• Boost (Program Options) for argument parsing
• Eigen3 for linear algebra

Note: Replace vcpkg paths in CMake if necessary (e.g., CMAKE_TOOLCHAIN_FILE).

1. Clone the repository:

   git clone https://github.com/helia77/tensorview3D.git
   cd tensorview3D

2. Create a build directory and configure CMake:

   mkdir build && cd build
   cmake -DCMAKE_TOOLCHAIN_FILE="path/to/vcpkg/scripts/buildsystems/vcpkg.cmake" ..

3. Build the project:

   cmake --build . --config Release

To install Python libraries, run on your conda terminal:

conda install numpy scikit-image scipy

• Build the C++ executable using CMake (steps above) and run:

  ./tensorview3 --input TENSOR_FILENAME

Contributions are welcome! If you'd like to improve this repository, follow these steps:

1. Fork the repository.
2. Create a feature branch:

   git checkout -b feature-branch

3. Commit your changes and push the branch.
4. Open a Pull Request.