Skip to content

This is the official implement for data generation of "Neural Volume Reconstruction for Coherent Synthetic Aperture Sonar" (SIGGRAPH 2023)

Notifications You must be signed in to change notification settings

juhyeonkim95/MitsubaPyOptiXTransient

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

46 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MitsubaPyOptiXTransient

This repository is the official data generation code for "Neural Volumetric Reconstructions for Coherent Synthetic Aperture Sonar" by Albert Reed, Juhyeon Kim, Thomas Blanford, Adithya Pediredla, Daniel C.Brown, Suren Jayasuriya (SIGGRAPH 2023).

Please also check the main code for the paper at here.

teasure

Renderer Overview

This is a custom python, GPU(OptiX)-based transient path tracing engine that renders Mitsuba formatted scenes which is built based on my previous custom renderer.

We implemented transient rendering using atomic operation on CUDA.

Compared to CPU transient rendering, GPU version is x10 to x50 faster.

Install

Please refer here to install the renderer.

Usage

python main_transient.py ../configs_transient_example/bunny.json

About

This is the official implement for data generation of "Neural Volume Reconstruction for Coherent Synthetic Aperture Sonar" (SIGGRAPH 2023)

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published