Software

Open-source tools built by the RISc Lab.

FalcorComp

GPU Monte Carlo renderers for computational imaging
The Cornell box rendered as a standard image, time-gated image, transient histogram, CW-ToF, structured light, Doppler spectrum, Doppler-gated image, Doppler ToF velocity, and events.

FalcorComp is a collection of rendering algorithms for computational imaging built on top of NVIDIA Falcor. It focuses on active sensing: time-of-flight, structured light, Doppler sensing, and event cameras. Every renderer runs on the GPU with hardware ray tracing, so it supports real-time, interactive simulation as well as offline rendering.

DocumentationGitHubPyPI

What it renders

Time-of-Flight

Time-gated images and transient histograms, with both path tracing and ReSTIR (spatio-temporal reservoir resampling).

Modulated Light

Continuous-wave ToF and structured light, with geometric antithetic sampling.

Doppler

Doppler-gated images, Doppler spectra, FMCW spectra, and Doppler ToF velocity.

Event Cameras

Brightness change and event generation, including the EventSVGF denoising pass.

Installation

pip install falcorcomp

Requires 64-bit Linux or Windows 11, an NVIDIA RTX GPU (hardware ray tracing), Python 3.9–3.13.

Papers implemented