Emission and Color Reproduction

We start by implementing the glow of perfectly diffuse emissive objects for some input temperature. In this case, our object is an ideal blackbody that absorbs and re-emits all energy falling upon it. After computing the radiance of our blackbody with Planck's Law of Blackbody Radiation, we convert this high dimensional spectrum to an sRGB color compatible with the computer display.

Reflectance and Polarization

We combined our estimate of blackbody radiation with the reflectance of the metal, allowing us to realistically render glow effects for various metallic materials. The final sampled color is a combination of reflected and emitted light, weighted according to Kirchhoff's Law of Thermal Radiation. We also explored how polarization of light changes glow effects and reflections.

Temperature Distributions

When raytracing, we sample the temperature at each point of intersection. These sampled temperatures are dependent on the type of temperature distribution given: uniform, linear, or noise-based. The linear distribution applies a smooth temperature gradient, and the noise map applies the pseudo-random Perlin noise function with control over temperature range and noise fineness.

Path Tracer Codebase

Our blackbody renderer was built as an extension of the CS184 Project 3: Pathtracer codebase. For this previous project, I implemented the core path tracing algorithms of a physically-based renderer, including ray-geometry intersection, direct and global illumination with Russian Roulette recursion, and adaptive sampling. Additionally, I developed microfacet material modeling and depth-of-field lens simulation.

For references, additional results, and information on our technical approach, see our report.