RGB-D mapping: Using Kinect-style depth cameras for dense 3D modeling of indoor environments

International Journal of Robotics Research - Tập 31 Số 5 - Trang 647-663 - 2012
Peter Henry1, Michael Krainin1, Evan Herbst1, Xiaofeng Ren2, Dieter Fox1
1Department of Computer Science and Engineering, University of Washington, Seattle, WA, USA
2ISTC-Pervasive Computing, Intel Labs, Seattle, WA, USA#TAB#

Tóm tắt

RGB-D cameras (such as the Microsoft Kinect) are novel sensing systems that capture RGB images along with per-pixel depth information. In this paper we investigate how such cameras can be used for building dense 3D maps of indoor environments. Such maps have applications in robot navigation, manipulation, semantic mapping, and telepresence. We present RGB-D Mapping, a full 3D mapping system that utilizes a novel joint optimization algorithm combining visual features and shape-based alignment. Visual and depth information are also combined for view-based loop-closure detection, followed by pose optimization to achieve globally consistent maps. We evaluate RGB-D Mapping on two large indoor environments, and show that it effectively combines the visual and shape information available from RGB-D cameras.

Từ khóa


Tài liệu tham khảo

10.1109/3DPVT.2006.141

10.1016/j.robot.2009.09.011

10.1109/34.121791

Bradski G (2000) The OpenCV Library. Dr. Dobb’s Journal of Software Tools. Available from: http://drdobbs.com/open-source/184404319

10.1007/978-3-540-88682-2_6

10.1016/0262-8856(92)90066-C

10.15607/RSS.2007.III.038

10.1109/CVPR.2010.5540082

10.1145/237170.237191

Droeschel D, 2009, European Conference on Mobile Robots (ECMR)

10.1145/2030112.2030123

10.1145/358669.358692

10.1109/TPAMI.2009.161

10.1109/ICCV.2009.5459145

10.1109/IROS.2007.4399030

10.15607/RSS.2007.III.009

Henry P, 2010, Proceedings of the International Symposium on Experimental Robotics (ISER)

10.1364/JOSAA.4.000629

10.1109/IM.1997.603871

10.1109/ICCVW.2009.5457430

Konolige K, 2004, Proceedings of the National Conference on Artificial Intelligence (AAAI)

10.1109/ROBOT.2010.5509796

10.5244/C.24.102

10.1109/TRO.2008.2004832

10.1177/0278364911403178

10.1145/1486525.1486527

10.1023/B:VISI.0000029664.99615.94

10.1023/A:1008854305733

10.1002/rob.20321

10.1109/CVPR.2010.5539794

10.1177/0278364909341483

10.1109/TPAMI.2004.17

10.1109/CVPR.2004.1315094

10.1109/CVPR.2006.264

10.1145/344779.344936

10.1007/s11263-007-0086-4

10.1504/IJISTA.2008.021298

10.15607/RSS.2007.III.026

10.1007/11744023_34

10.1109/ICRA.2011.5980220

10.1109/IM.2001.924423

Se S, 2008, International Journal of Intelligent Control and Systems, 13, 47

10.15607/RSS.2009.V.021

10.1109/34.982886

10.1109/CVPR.2011.5995316

10.1145/1141911.1141964

10.1109/IROS.2009.5354708

10.1109/ROBOT.2000.844077

Thrun S, 2005, Probabilistic Robotics

10.1006/cviu.1999.0832

Triebel R, 2005, Proceedings of the National Conference on Artificial Intelligence (AAAI)

Triggs B, 2000, Vision Algorithms: Theory and Practice, 1883, 153

Wu C. (2007). SiftGPU: A GPU implementation of scale invariant feature transform (SIFT). Available at: http://cs.unc.edu/~ccwu/siftgpu.

Thông tin
Thông tin xuất bản

International Journal of Robotics Research

Tập 31 Số 5

647-663

Thông tin tác giả