3D Ray Tracing Engine (Project CSI - Visual and Analytical Computing)

A high-performance Monte Carlo Path Tracer with an interactive GUI, built from scratch in C++17. This project implements physically-based rendering (PBR) techniques including GGX microfacet materials, Bounding Volume Hierarchy (BVH) acceleration, and AI-powered denoising.

🎯 Features

Rendering

• Monte Carlo Path Tracing with unbiased global illumination
• Russian Roulette path termination for efficiency
• Multiple Importance Sampling (MIS) with power heuristic
• Next Event Estimation (NEE) for direct light sampling

Materials (PBR)

• Lambertian diffuse surfaces
• Metal with configurable roughness
• Dielectric (glass/water) with Fresnel and refraction
• GGX Microfacet for realistic metallic highlights
• Emissive materials for area lights

Acceleration & Optimization

• BVH (Bounding Volume Hierarchy) with Surface Area Heuristic
• Multi-threaded tile-based rendering using C++ threads
• Intel OIDN AI denoiser integration

Interactive GUI

• Real-time progressive rendering
• WASD + mouse camera controls
• Live material and lighting adjustments
• Scene file hot-reloading

---

� Documentation

For a comprehensive technical documentation covering all algorithms, mathematical foundations, and implementation details, see:

📄 Project_Documentation.pdf - Complete 13-page technical report

The documentation includes:

• Rendering equation and Monte Carlo integration theory
• Ray-scene intersection algorithms (Möller-Trumbore)
• BVH construction with Surface Area Heuristic
• Material models (Lambertian, GGX, Dielectric)
• Importance sampling techniques (MIS, NEE, Russian Roulette)
• System architecture and performance analysis

---

�🛠️ Tools & Technologies

Category	Technology
Language	C++17
Build System	CMake 3.16+
GUI	Dear ImGui + GLFW + OpenGL
Math	GLM
Scene Format	XML (tinyxml2) + OBJ/MTL
Denoising	Intel Open Image Denoiser (OIDN)
Image I/O	stb_image, stb_image_write

---

📋 Prerequisites

Ensure you have the following installed:

• CMake 3.16 or higher
• C++17 compatible compiler (GCC 9+, Clang 10+, MSVC 2019+)
• OpenGL 3.3+ compatible GPU and drivers
• Git (for cloning)

macOS

brew install cmake glfw

Ubuntu/Debian

sudo apt update
sudo apt install cmake build-essential libglfw3-dev libgl1-mesa-dev

Windows

• Install Visual Studio 2019+ with C++ workload
• Install CMake

---

🚀 Build Instructions

1. Clone the Repository

git clone https://github.com/YOUR_USERNAME/project-csi.git
cd project-csi

2. Create Build Directory

mkdir build
cd build

3. Configure with CMake

cmake ..

4. Build the Project

cmake --build . --config Release -j$(nproc)

Or on Windows:

cmake --build . --config Release

---

▶️ Running the Application

Interactive GUI Mode

# From the build directory
./RaytracerUI

Command-Line Renderer

# Render with default settings
./Raytracer

# Render a specific scene
./Raytracer --scene cornell_water_scene.xml --bvh --samples 500

# Full options
./Raytracer --help

CLI Options

Flag	Description
--scene <file>	Scene XML file (default: objects.xml)
--out <file>	Output image path
--width <W>	Override image width
--samples <S>	Samples per pixel
--bvh	Use BVH acceleration (recommended)
--denoise	Enable AI denoising (default)
--preset <name>	Use preset: Preview, Draft, Final

---

📁 Project Structure

project-csi/
├── assets/                 # Scene files and 3D models
│   ├── *.xml              # Scene definitions
│   ├── *.obj              # 3D mesh files
│   └── *.mtl              # Material files
├── src/
│   ├── engine/            # Core ray tracing engine
│   │   ├── camera.h/cpp   # Camera model
│   │   ├── material.h     # Material base class
│   │   ├── bvh_node.h/cpp # BVH acceleration
│   │   ├── render_runner.cpp # Tile-based renderer
│   │   └── ...
│   ├── gui/               # ImGui application
│   └── 3rdParty/          # External libraries
├── renders/samples/       # Sample output images
├── build/output/          # Rendered output directory
├── CMakeLists.txt
└── README.md

---

🖼️ Sample Outputs

Hundred Spheres with Sharp Shadowing and Lighting using BVH

Demonstrates refraction, caustics, and global illumination

GGX Material Test

Showcases metallic roughness variations

100 Spheres

BVH acceleration test with multiple objects

Random Scene

Scene with Chair Obj Rendering and Point Light Material

---

🎬 Demo Video

Interactive GUI Demo

test_video1.mp4

Real-time Rendering Demo

test_video3.mp4

Demonstration videos showing the interactive GUI and real-time rendering capabilities.

---

📖 Key Concepts Implemented

The Rendering Equation

Lo(x,ωo) = Le(x,ωo) + ∫ f(x,ωi,ωo) Li(x,ωi) cos(θi) dωi

BVH Acceleration

Reduces ray-scene intersection from O(n) to O(log n) using:

• Axis-Aligned Bounding Boxes (AABB)
• Surface Area Heuristic (SAH) for optimal tree construction

GGX Microfacet Model

Physically accurate BRDF with:

• Normal Distribution Function (NDF)
• Smith Geometry Function
• Schlick Fresnel Approximation

---

👥 Authors

• Abu Bakar

Supervisor: Professor Olivier Staadt
Institution: University of Rostock
Course: Project CSI - Visual and Analytical Computing (12 ECTS)

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

🙏 Acknowledgements

• Ray Tracing in One Weekend by Peter Shirley
• Intel Open Image Denoise
• Dear ImGui
• GLFW
• GLM