A low-cost stereo system for 3D object recognition

In this paper, we present a low-cost stereo vision system designed for object recognition with FPFH point feature descriptors. Image acquisition is performed using a pair of consumer market UVC cameras costing less than 80 Euros, lacking synchronization signal and without customizable optics. Nonetheless, the acquired point clouds are sufficiently accurate to perform object recognition using FPFH features. The recognition algorithm compares the point cluster extracted from the current image pair with the models contained in a dataset. Experiments show that the recognition rate is above 80% even when the object is partially occluded.

[1]  Kurt Konolige,et al.  Projected texture stereo , 2010, 2010 IEEE International Conference on Robotics and Automation.

[2]  Andrew E. Johnson,et al.  Spin-Images: A Representation for 3-D Surface Matching , 1997 .

[3]  Dongseok Ryu,et al.  Multiple Intensity Differentiation for 3-D Surface Reconstruction With Mono-Vision Infrared Proximity Array Sensor , 2011, IEEE Sensors Journal.

[4]  Hirokazu Kato,et al.  Marker tracking and HMD calibration for a video-based augmented reality conferencing system , 1999, Proceedings 2nd IEEE and ACM International Workshop on Augmented Reality (IWAR'99).

[5]  Stefano Caselli,et al.  Object categorization and grasping by parts from range scan data , 2012, 2012 IEEE International Conference on Robotics and Automation.

[6]  Zoltan-Csaba Marton,et al.  Tutorial: Point Cloud Library: Three-Dimensional Object Recognition and 6 DOF Pose Estimation , 2012, IEEE Robotics & Automation Magazine.

[7]  Michael Riis Andersen,et al.  Kinect Depth Sensor Evaluation for Computer Vision Applications , 2012 .

[8]  Radu Bogdan Rusu,et al.  3D is here: Point Cloud Library (PCL) , 2011, 2011 IEEE International Conference on Robotics and Automation.

[9]  Wolfram Burgard,et al.  Point feature extraction on 3D range scans taking into account object boundaries , 2011, 2011 IEEE International Conference on Robotics and Automation.

[10]  Matthijs C. Dorst Distinctive Image Features from Scale-Invariant Keypoints , 2011 .

[11]  Nico Blodow,et al.  Fast Point Feature Histograms (FPFH) for 3D registration , 2009, 2009 IEEE International Conference on Robotics and Automation.

[12]  Gilles Burel,et al.  Three-dimensional invariants and their application to object recognition , 1995, Signal Process..

[13]  Michael Garland,et al.  Curvature maps for local shape comparison , 2005, International Conference on Shape Modeling and Applications 2005 (SMI' 05).