Object Recognition Using Image Warping in an Intelligent Space

We propose a method to improve object recognition performance of a robot with an intelligent space system (iSpace). Existing SIFT-based object recognition is powerful, but it has still some limitations. Among limitations, we focus on degraded performance according to the viewpoint change. In order to improve the performance, we use appearance estimation of the object in the point of robot's view using image warping based on the robot pose and object pose. The proposed method is evaluated by experiments. 

[1]  David Nistér,et al.  Scalable Recognition with a Vocabulary Tree , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[2]  Janne Heikkilä,et al.  Geometric Camera Calibration Using Circular Control Points , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Jean-Michel Morel,et al.  ASIFT: A New Framework for Fully Affine Invariant Image Comparison , 2009, SIAM J. Imaging Sci..

[4]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[5]  Yan Ke,et al.  PCA-SIFT: a more distinctive representation for local image descriptors , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[6]  M. Turk,et al.  Eigenfaces for Recognition , 1991, Journal of Cognitive Neuroscience.

[7]  Luc Van Gool,et al.  Speeded-Up Robust Features (SURF) , 2008, Comput. Vis. Image Underst..

[8]  Changchang Wu,et al.  SiftGPU : A GPU Implementation of Scale Invariant Feature Transform (SIFT) , 2007 .

[9]  Yan Ke,et al.  PCA-SIFT: a more distinctive representation for local image descriptors , 2004, CVPR 2004.

[10]  Cordelia Schmid,et al.  A performance evaluation of local descriptors , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).