ロボット3D物体分類におけるSim2Realギャップの解消
Addressing the Sim2Real Gap in Robotic 3D Object Classification
CADモデルで学習し実環境の再構成物体で評価する際のギャップに対し、変換に頑健な物体パーツを抽出しグラフ畳み込みで分類する手法を提案した。
著者: Jean-Baptiste Weibel, Timothy Patten, Markus Vincze
分類: cs.CV, cs.RO
原文アブストラクト
Object classification with 3D data is an essential component of any scene understanding method. It has gained significant interest in a variety of communities, most notably in robotics and computer graphics. While the advent of deep learning has progressed the field of 3D object classification, most work using this data type are solely evaluated on CAD model datasets. Consequently, current work does not address the discrepancies existing between real and artificial data. In this work, we examine this gap in a robotic context by specifically addressing the problem of classification when transferring from artificial CAD models to real reconstructed objects. This is performed by training on ModelNet (CAD models) and evaluating on ScanNet (reconstructed objects). We show that standard methods do not perform well in this task. We thus introduce a method that carefully samples object parts that are reproducible under various transformations and hence robust. Using graph convolution to classify the composed graph of parts, our method significantly improves upon the baseline.
