Exploiting Color Constancy for Robust Tracking Under Non-uniform Illumination

Tracking objects in environments under non-uniform illumination condition is particularly challenging as the observed appearance may change in space and time. Adapting the appearance model increases the risk of drifts, while iltering out the illumination information through built-in invariance reduces the discriminative capabilities. In this work we adhere to color constancy principles to learn the appearance variation induced by non-uniform illumination and we use this information to perform location-dependent color corrections to boost tracking performance. The training procedure is carried out in an unsupervised manner by exploiting walking people as illumination probes and an on-line, non-parametric regression method is developed to densely predict the location-specific color transformations.

[1]  Elisa Ricci,et al.  Tracking Multiple People with Illumination Maps , 2010, 2010 20th International Conference on Pattern Recognition.

[2]  Thierry Chateau,et al.  Illumination aware MCMC Particle Filter for long-term outdoor multi-object simultaneous tracking and classification , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[3]  Luis Salgado,et al.  Adaptive background modeling in multicamera system for real-time object detection , 2011 .

[4]  David J. Fleet,et al.  Robust Online Appearance Models for Visual Tracking , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Rainer Stiefelhagen,et al.  Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics , 2008, EURASIP J. Image Video Process..

[6]  Tao Hu,et al.  Learning the Scene Illumination for Color-Based People Tracking in Dynamic Environment , 2013, ICIAP.

[7]  Vittorio Murino,et al.  Background Subtraction for Automated Multisensor Surveillance: A Comprehensive Review , 2010, EURASIP J. Adv. Signal Process..

[8]  Ming-Hsuan Yang,et al.  Visual tracking with online Multiple Instance Learning , 2009, CVPR.

[9]  Mark S. Drew,et al.  Diagonal transforms suffice for color constancy , 1993, 1993 (4th) International Conference on Computer Vision.

[10]  Oswald Lanz,et al.  Approximate Bayesian multibody tracking , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Joost van de Weijer,et al.  Computational Color Constancy: Survey and Experiments , 2011, IEEE Transactions on Image Processing.

[12]  Stefano Messelodi,et al.  Linking the von Kries model to Wien's law for the estimation of an illuminant invariant image , 2011, Pattern Recognit. Lett..

[13]  Thomas S. Huang,et al.  Image processing , 1971 .

[14]  Oswald Lanz,et al.  A Sampling Algorithm for Occlusion Robust Multi Target Detection , 2009, 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance.

[15]  Jiri Matas,et al.  Linear Regression and Adaptive Appearance Models for Fast Simultaneous Modelling and Tracking , 2010, International Journal of Computer Vision.

[16]  Ming-Hsuan Yang,et al.  Robust Object Tracking with Online Multiple Instance Learning , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[17]  Keigo Hirakawa,et al.  Color Constancy with Spatio-Spectral Statistics , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Yi-Ping Hung,et al.  Background Removal of Multiview Images by Learning Shape Priors , 2007, IEEE Transactions on Image Processing.

[19]  Rita Cucchiara,et al.  Probabilistic people tracking with appearance models and occlusion classification: The AD-HOC system , 2011, Pattern Recognit. Lett..