Online learning for automatic segmentation of 3D data

Abstract — We propose a method to perform automatic seg-mentation of 3D scenes based on a standard classifier, whose learning model is continuously improved by means of new samples, and a grouping stage, that enforces local consistency among classified labels. The new samples are automatically delivered to the system by a feedback loop based on a feature selection approach that exploits the outcome of the grouping stage. By experimental results on several datasets we demon-strate that the proposed online learning paradigm is effective in increasing the accuracy of the whole 3D segmentation thanks to the improvement of the learning model of the classifier by means of newly acquired, unsupervised data. I. INTRODUCTION AND PREVIOUS WORK Fostered by the increasing availability of accurate and low-cost 3D sensors, a relevant research topic within th