Fast Pixelwise Adaptive Visual Tracking of Non-Rigid Objects

• Published in: IEEE transactions on image processing Vol. 26; no. 5; pp. 2368 - 2380
• Main Authors: Duffner, Stefan, Garcia, Christophe
• Format: Journal Article
• Language: English
• Published: United States IEEE 01-05-2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE); Institute of Electrical and Electronics Engineers
• Subjects: Adaptation models; Adaptive algorithms; Algorithms; Artificial Intelligence; Color; Computational modeling; Computer Science; Computer Vision and Pattern Recognition; Deformation effects; Document and Text Processing; Hough transformation; image motion analysis; Image Processing; Image segmentation; Image sequence analysis; Machine Learning; Multimedia; Neural and Evolutionary Computing; Optical tracking; Probabilistic methods; Real time; Robustness; Segmentation; Shape; Signal and Image Processing; State-of-the-art reviews; Tracking; Videos; Adaptation models; Image segmentation; Shape; Tracking; Computational modeling; Robustness; Videos
• ISSN: 1057-7149; 1941-0042
• DOI: 10.1109/TIP.2017.2676346

In this paper, we present a new algorithm for real-time single-object tracking in videos in unconstrained environments. The algorithm comprises two different components that are trained "in one shot" at the first video frame: a detector that makes use of the generalized Hough transform with color and gradient descriptors and a probabilistic segmentation method based on global models for foreground and background color distributions. Both components work at pixel level and are used for tracking in a combined way adapting each other in a co-training manner. Moreover, we propose an adaptive shape model as well as a new probabilistic method for updating the scale of the tracker. Through effective model adaptation and segmentation, the algorithm is able to track objects that undergo rigid and non-rigid deformations and considerable shape and appearance variations. The proposed tracking method has been thoroughly evaluated on challenging benchmarks, and outperforms the state-of-the-art tracking methods designed for the same task. Finally, a very efficient implementation of the proposed models allows for extremely fast tracking.