Non-Uniform Integration of TDCI Captures

Paul Eberhart and Henry Dietz

Department of Electrical and Computer Engineering
Center for Visualization & Virtual Environments
University of Kentucky, Lexington, KY 40506-0046

Original March 22, 2020, Latest Update March 22, 2020

This document should be cited using something like the bibtex entry:

author={Paul Eberhart},
author={Henry Dietz},
title={{Non-Uniform Integration of TDCI Captures}},
institution={University of Kentucky},
howpublished={Aggregate.Org online technical report},


We have published quite a few things on TDCI -- Time Domain Continuous Imaging. This page is about the most recent: "Non-Uniform Integration of TDCI Captures," by Paul Eberhart and Henry Dietz, which was a poster presentation at IS&T Electronic Imaging 2020. A preprint version is posted here. The poster for this paper is also posted here.

Prototype Implementation

There are two pieces of code associated with this paper.

The first is an Octave script, NonUniformPoC.m, which depends only on FFMPEG. This script can ingest a sequence of video frames to simulate continuously sampled incident light, and apply user-specified spatial and temporal non-uniformities to the integration of that light. This very crude proof-of-concept implementation serves primarily as a testbed for algorithms and representations, as well as an easy way to experiment with the effects which can be produced, and is not tuned to be particularly fast or high quality.

The second is FnSubTest.c, which uses libnetpbm. This is isolated code to read functions and masks, generating a gain mask. Unlike the Octave code, this is not a complete implementation of the processing, but just a key step.

The Aggregate. The only thing set in stone is our name.