Reza Farid, Region-Growing Planar Segmentation for Robot Action Planning, In Bernhard Pfahringer and Jochen Renz, editors, AI2015: Advances in Artificial Intelligence, volume 9457 of Lecture Notes in Artificial Intelligence, pp. 179-191. Springer International Publishing, 2015, doi: 10.1007/978-3-319-26350-2_16. [Code & Data]
Abstract: Object detection, classification and manipulation are some of the capabilities required by autonomous robots. The main steps in object classification are: segmentation, feature extraction, object representation and learning. To address the problem of learning object classification using multi-view range data, we used a relational approach. The first step of our object classification method is to decompose a scene into shape primitives such as planes, followed by extracting a set of higher-level, relational features from the segmented regions. In this paper, we compare our plane segmentation algorithm with state-of-the-art plane segmentation algorithms which are publicly available. We show that our segmentation outperforms visually and also produces better results for the robot action planning. Keywords: object classification, robot action planning, planar segmentation, point cloud, range data.