Journal
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
Volume 17, Issue 1, Pages 388-397Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TII.2020.2986086
Keywords
Cameras; Robot vision systems; Informatics; Training; Three-dimensional displays; Image recognition; CNN; distribution; Gaussian mixture; mixture density; particle filter
Categories
Funding
- Industry Core Technology Development Project - Ministry of Trade, Industry & Energy (MOTIE, South Korea) [20005062]
Ask authors/readers for more resources
This article introduces a new deep neural network named MD-PoseNet to tackle the challenging problem of global localization using a monocular camera. Unlike existing methods, MD-PoseNet returns multiple guesses for the camera pose, which are then exploited in the probabilistic framework of particle filters.
Global localization using a monocular camera is one of the most challenging problems in computer vision and intelligent robotics. In this article, a new deep neural network named Mixture Density (MD)-PoseNet is proposed to address this problem. Unlike existing learning-based global localization methods that return a single guess for the camera pose, MD-PoseNet returns multiple guesses represented in the form of a Gaussian mixture (GM). The key idea of MD-PoseNet is that the network returns the distribution of all probable camera poses instead of the most probable camera pose, and the distribution represents the multiple guesses for the camera pose. The multiple guesses returned by MD-PoseNet are, consequently, exploited in the probabilistic framework of particle filters. Finally, the proposed method is applied to four different environments, and its validity is demonstrated via experiments.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available