Title :
Self-Paced Learning for Long-Term Tracking
Author :
Supancic, James Steven ; Ramanan, D.
Author_Institution :
Dept. of Comput. Sci., Univ. of California, Irvine, Irvine, CA, USA
Abstract :
We address the problem of long-term object tracking, where the object may become occluded or leave-the-view. In this setting, we show that an accurate appearance model is considerably more effective than a strong motion model. We develop simple but effective algorithms that alternate between tracking and learning a good appearance model given a track. We show that it is crucial to learn from the "right" frames, and use the formalism of self-paced curriculum learning to automatically select such frames. We leverage techniques from object detection for learning accurate appearance-based templates, demonstrating the importance of using a large negative training set (typically not used for tracking). We describe both an offline algorithm (that processes frames in batch) and a linear-time on-line (i.e. causal) algorithm that approaches real-time performance. Our models significantly outperform prior art, reducing the average error on benchmark videos by a factor of 4.
Keywords :
learning (artificial intelligence); object detection; object tracking; appearance-based templates; good appearance model; large negative training set; linear-time on-line algorithm; long-term object tracking; object detection; offline algorithm; self-paced curriculum learning; Adaptation models; Computational modeling; Detectors; Real-time systems; Support vector machines; Training; Videos; learning; object tracking; self paced learning; tracking;
Conference_Titel :
Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on
Conference_Location :
Portland, OR
DOI :
10.1109/CVPR.2013.308