Visual information abstraction for interactive robot learning

Abstract

Semantic visual perception for knowledge acquisition plays an important role in human cognition, as well as in the learning process of any cognitive robot. In this paper, we present a visual information abstraction mechanism designed for continuously learning robotic systems. We generate spatial information in the scene by considering plane estimation and stereo line detection coherently within a unified probabilistic framework, and show how spaces of interest (SOIs) are generated and segmented using the spatial information. We also demonstrate how the existence of SOIs is validated in the long-term learning process. The proposed mechanism facilitates robust visual information abstraction which is a requirement for continuous interactive learning. Experiments demonstrate that with the refined spatial information, our approach provides accurate and plausible representation of visual objects.

Spatial structure analysis for autonomous robotic vision systems

“…The mechanisms for analysis of 3D spatial structures have also been updated from that used for traditional static scenes (e.g. detecting the dominant plane in the scene for enabling the robot to grasp the objects which pop out from the planar surface [12] or to learn the properties of the objects with respect to [21]) challenging dynamic scenes (e.g. search multiple objects on various supporting surfaces [14][1]).…”

Robot George - interactive continuous learning of visual concepts

“…Given that the system learns from a variety of as yet unknown objects, we implemented a generic active object detection and segmentation scheme that uses 2D and 3D information, exploiting the fact that objects are presented on planar supporting surfaces [1].…”