# diff-triangle-tracer

**Repository Path**: yejun668/diff-triangle-tracer

## Basic Information

- **Project Name**: diff-triangle-tracer
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-01-20
- **Last Updated**: 2026-01-20

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Differentiable Triangle Tracer

This repository contains the OptiX implementation of the Differentiable Triangle Tracer used in UTrice: Unifying Primitives in Differentiable Ray Tracing and Rasterization via Triangles for Particle-Based 3D Scenes.

## Compilation

The tracer can be built using either the Conda toolchain or a manual setup. For the Conda workflow, please refer to the instructions in [UTrice](https://github.com/waizui/UTrice).

For manual compilation, you need to provide the libTorch include directories and libraries in CMakeLists.txt. Replace the following paths with your own configuration:

```bash
# Use torch from conda, python3.11 only
set(TORCH_ROOT "$ENV{CONDA_PREFIX}/lib/python3.11/site-packages/torch")
set(TORCH_INCLUDE_DIRS
    ${TORCH_ROOT}/include
    ${TORCH_ROOT}/include/torch/csrc/api/include
)
set(TORCH_LIBRARIES
    ${TORCH_ROOT}/lib/libtorch_python.so
    ${TORCH_ROOT}/lib/libtorch_cuda.so
    ${TORCH_ROOT}/lib/libtorch_cpu.so
    ${TORCH_ROOT}/lib/libc10_cuda.so
    ${TORCH_ROOT}/lib/libc10.so
)
```

Then run build.sh.

## Citation
```
@misc{liu2025utriceunifyingprimitivesdifferentiable,
      title={UTrice: Unifying Primitives in Differentiable Ray Tracing and Rasterization via Triangles for Particle-Based 3D Scenes}, 
      author={Changhe Liu and Ehsan Javanmardi and Naren Bao and Alex Orsholits and Manabu Tsukada},
      year={2025},
      eprint={2512.04421},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2512.04421}, 
}
```