Efficient Selection of Disambiguating Actions for Stereo Vision

TL;DR

This paper presents an efficient method for selecting disambiguating actions in stereo vision, using structured models to reduce computation time for active sensing in robotics.

Contribution

It introduces a novel approach to choose laser disambiguation points in stereo vision with reduced computational complexity, enabling real-time active sensing.

Findings

Entropy minimizing laser points can be computed in O(nd) time

The method outperforms typical HMM-based approaches in computational efficiency

Applicable to scenes with weak texture where stereo matching is challenging

Abstract

In many domains that involve the use of sensors, such as robotics or sensor networks, there are opportunities to use some form of active sensing to disambiguate data from noisy or unreliable sensors. These disambiguating actions typically take time and expend energy. One way to choose the next disambiguating action is to select the action with the greatest expected entropy reduction, or information gain. In this work, we consider active sensing in aid of stereo vision for robotics. Stereo vision is a powerful sensing technique for mobile robots, but it can fail in scenes that lack strong texture. In such cases, a structured light source, such as vertical laser line can be used for disambiguation. By treating the stereo matching problem as a specially structured HMM-like graphical model, we demonstrate that for a scan line with n columns and maximum stereo disparity d, the entropy minimizing aim point for the laser can be selected in O(nd) time - cost no greater than the stereo algorithm itself. In contrast, a typical HMM formulation would suggest at least O(nd^2) time for the entropy calculation alone.