A Spatio-Temporal
Probabilistic Model for Multi-Sensor Multi-Class Object
Recognition
Proc. of ISRR, 2007
Abstract
Abstract.This paper presents a general probabilistic
framework for multi- sensor multi-class ob ject recognition based on
Conditional Random Fields (CRFs) trained with virtual evidence
boosting. The learnt representation models spatial and temporal
relationships and is able to integrate arbitrary sensor information
by automatically extracting features from data. We demonstrate the
benefits of modelling spatial and temporal relationships for the
problem of detecting seven classes of ob jects using laser and
vision data in outdoor environments. Additionally, we show how
this framework can be used with partially labeled data, thereby
significantly reducing the burden of manual data annotation.