Name
Abstract | ||
|---|---|---|
In many domains that involve the use of sensors, such as robotics or sensor networks, there are op- portunities to use some form of active sensing to disambiguate data from noisy or unreliable sen- sors. 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 ac- tive 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 match- ing problem as a specially structured HMM-like graphical model, we demonstrate that for a scan line with n columns and maximum stereo dis- parity d, the entropy minimizing aim point for the laser can be selected in O(nd) time - cost no greater than the stereo algorithm itself. A typical HMM formulation would suggest at least O(nd2) time for the entropy calculation alone. |
Year | Venue | Keywords |
|---|---|---|
2006 | Uncertainty in Artificial Intelligence | information gain,mobile robot,graphical model,sensor network,stereo vision,structured light |
Field | DocType | Volume |
Stereo cameras,Stereopsis,Computer science,Artificial intelligence,Robotics,Scan line,Computer vision,Structured light,Pattern recognition,Graphical model,Machine learning,Mobile robot,Computer stereo vision | Conference | abs/1206.6878 |
Citations | PageRank | References |
0 | 0.34 | 9 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Monika Schaeffer | 1 | 0 | 0.34 |
| Ronald Parr | 2 | 2428 | 186.85 |