A nonparametric approach to region-of-interest detection in wide-angle views

作者：Cheng Guangchun^*; Buckles Bill P

来源：Pattern Recognition Letters, 2014, 49: 24-32.

DOI：10.1016/j.patrec.2014.05.019

• Region of interest
• Anomaly detection
• Structure tensor
• Unsupervised learning
• Video analytics

We propose a tracking-free method to detect the regions of interest (ROI) in a wide-angle video stream. A region is defined as a statistical outlier among occurrences of motion patterns, and is detected in an unsupervised manner. Based on 3D structure tensors, the activity at any site is modeled by the probability distribution of distances between structure tensors. The distribution is estimated using a nonparametric kernel density estimator. The detection of regions is determined by observing a long period of low-probability motion occurrences. Experiments performed with real-world datasets indicate that the proposed algorithm can detect both spatial ROIs and spatio-temporal ROIs, and outperforms other nonparametric methods. Published by Elsevier B.V.

• 出版日期2014-11-1