Viewpoint Evaluation for Online 3-D Active Object Classification

被引：34

作者：

Patten, Timothy ^{[1
]}

Zillich, Michael ^{[2
]}

Fitch, Robert ^{[1
]}

Vincze, Markus ^{[2
]}

Sukkarieh, Salah ^{[1
]}

机构：

[1] Univ Sydney, Australian Ctr Field Robot, Sydney, NSW, Australia

[2] Vienna Univ Technol, Automat & Control Inst, Vision4Robot Grp, Vienna, Austria

来源：

IEEE ROBOTICS AND AUTOMATION LETTERS | 2016年 / 1卷 / 01期

基金：

澳大利亚研究理事会;

关键词：

Object detection; segmentation; categorization; Semantic scene understanding; RGB-D perception;

D O I：

10.1109/LRA.2015.2506901

中图分类号：

TP24 [机器人技术];

学科分类号：

080202 ; 1405 ;

摘要：

We present an end-to-end method for active object classification in cluttered scenes from RGB-D data. Our algorithms predict the quality of future viewpoints in the form of entropy using both class and pose. Occlusions are explicitly modeled in predicting the visible regions of objects, which modulates the corresponding discriminatory value of a given view. We implement a one-step greedy planner and demonstrate our method online using a mobile robot. We also analyze the performance of our method compared to similar strategies in simulated execution using the Willow Garage dataset. Results show that our active method usefully reduces the number of views required to accurately classify objects in clutter as compared to traditional passive perception.

引用

页码：73 / 81

页数：9

共 50 条

[1] Toward Online 3-D Object Segmentation and Mapping
Herbst, Evan
Henry, Peter
Fox, Dieter
2014 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2014, : 3193 - 3200
[2] Adaptive 3-D Object Classification with Reinforcement Learning
Garstka, Jens
Peters, Gabriele
ICIMCO 2015 PROCEEDINGS OF THE 12TH INTERNATIONAL CONFERENCE ON INFORMATICS IN CONTROL, AUTOMATION AND ROBOTICS, VOL. 2, 2015, : 381 - 385
[3] Evaluation of Local 3-D Point Cloud Descriptors in Terms of Suitability for Object Classification
Garstka, Jens
Peters, Gabriele
ICINCO: PROCEEDINGS OF THE 13TH INTERNATIONAL CONFERENCE ON INFORMATICS IN CONTROL, AUTOMATION AND ROBOTICS, VOL 2, 2016, : 540 - 547
[4] Information entropy-based viewpoint planning for 3-D object reconstruction
Li, YF
Liu, ZG
IEEE TRANSACTIONS ON ROBOTICS, 2005, 21 (03) : 324 - 337
[5] Automatic class selection and prototyping for 3-D object classification
Donamukkala, R
Huber, D
Kapuria, A
Hebert, M
Fifth International Conference on 3-D Digital Imaging and Modeling, Proceedings, 2005, : 64 - 71
[6] Object segmentation and classification using 3-D range camera
Wei, Xue
Phung, Son Lam
Bouzerdoum, Abdesselam
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2014, 25 (01) : 74 - 85
[7] Active visual sensing of the 3-D pose of a flexible object
Byun, JE
Nagata, T
ROBOTICA, 1996, 14 : 173 - 188
[8] Optoelectronic 3-D Object Classification From 2-D Images
Loo, Chye-Hwa
Elsherbeni, Atef Z.
JOURNAL OF LIGHTWAVE TECHNOLOGY, 2008, 26 (17-20) : 3248 - 3255
[9] Viewpoint-dependent versus -independent 3-D object perception: A direct comparison
Vanrie, J
Wagemans, J
PSYCHOLOGICA BELGICA, 2001, 41 (03) : 115 - 129
[10] A Convolutional Learning System for Object Classification in 3-D Lidar Data
Prokhorov, Danil
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2010, 21 (05): : 858 - 863

← 1 2 3 4 5 →