A novel stereo matching approach for pedestrian re-identification

Automatic and reliable identification of pedestrians from multiple camera views is very important for video surveillance and can save a lot of manual effort. The significant variations in viewpoints, poses, illumination and occlusions makes this problem very challenging. Most of the existing approaches addressing this problem handle drastic viewpoint change in a supervised way and thus require labelling new training data for a different pair of camera views. In this paper, we present a novel approach for pedestrian re-identification using stereo matching, which does not require any kind of training. The cost of the stereo matching of two images is used for evaluating the similarity of the images, without performing 3-D reconstruction. We show that this cost is robust to the large pose variations observed in the images captured from multiple cameras. The proposed pedestrian re-identification algorithm is built on top of a dynamic programming stereo matching algorithm. Experimental evaluation on the challenging VIPeR dataset shows the effectiveness of the proposed approach.

[1]  Bingpeng Ma,et al.  BiCov: a novel image representation for person re-identification and face verification , 2012, BMVC.

[2]  Richard I. Hartley,et al.  Person Reidentification Using Spatiotemporal Appearance , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[3]  Alessandro Perina,et al.  Person re-identification by symmetry-driven accumulation of local features , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  Vittorio Murino,et al.  Custom Pictorial Structures for Re-identification , 2011, BMVC.

[5]  Xiaogang Wang,et al.  Locally Aligned Feature Transforms across Views , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Hai Tao,et al.  Evaluating Appearance Models for Recognition, Reacquisition, and Tracking , 2007 .

[7]  Bingpeng Ma,et al.  Local Descriptors Encoded by Fisher Vectors for Person Re-identification , 2012, ECCV Workshops.

[8]  Horst Bischof,et al.  Relaxed Pairwise Learned Metric for Person Re-identification , 2012, ECCV.

[9]  Hai Tao,et al.  Viewpoint Invariant Pedestrian Recognition with an Ensemble of Localized Features , 2008, ECCV.

[10]  Carlos D. Castillo,et al.  Using Stereo Matching with General Epipolar Geometry for 2D Face Recognition across Pose , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Frédéric Jurie,et al.  PCCA: A new approach for distance learning from sparse pairwise constraints , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[12]  Xiaogang Wang,et al.  Shape and Appearance Context Modeling , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[13]  Shaogang Gong,et al.  Person Re-Identification by Support Vector Ranking , 2010, BMVC.

[14]  Chunxiao Liu,et al.  Person Re-identification: What Features Are Important? , 2012, ECCV Workshops.

[15]  Andrew Blake,et al.  Efficient Dense Stereo with Occlusions for New View-Synthesis by Four-State Dynamic Programming , 2006, International Journal of Computer Vision.

[16]  Xiaogang Wang,et al.  Unsupervised Salience Learning for Person Re-identification , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

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

[18]  Shaogang Gong,et al.  Person re-identification by probabilistic relative distance comparison , 2011, CVPR 2011.