MVTrans: 透明物体のマルチビュー知覚
MVTrans: Multi-View Perception of Transparent Objects
RGB-Dの深度に頼らず、ステレオを拡張したマルチビュー入力で透明物体の深度推定・セグメンテーション・姿勢推定を同時に行う手法を提案し、大規模合成データセットSyn-TODDも構築した。
著者: Yi Ru Wang, Yuchi Zhao, Haoping Xu, Saggi Eppel, Alan Aspuru-Guzik, Florian Shkurti, Animesh Garg
分類: cs.RO, cs.AI, cs.CV
原文アブストラクト
Transparent object perception is a crucial skill for applications such as robot manipulation in household and laboratory settings. Existing methods utilize RGB-D or stereo inputs to handle a subset of perception tasks including depth and pose estimation. However, transparent object perception remains to be an open problem. In this paper, we forgo the unreliable depth map from RGB-D sensors and extend the stereo based method. Our proposed method, MVTrans, is an end-to-end multi-view architecture with multiple perception capabilities, including depth estimation, segmentation, and pose estimation. Additionally, we establish a novel procedural photo-realistic dataset generation pipeline and create a large-scale transparent object detection dataset, Syn-TODD, which is suitable for training networks with all three modalities, RGB-D, stereo and multi-view RGB. Project Site: https://ac-rad.github.io/MVTrans/
