Abstract
We propose an original method for detecting and localizing anomalous motion patterns in videos from a camera view -based motion representation perspective. Anomalous motion should be taken in a broad sense, i.e., unexpected, abnormal, singular, irregular, or unusual motion. Identifying distinctive dynamic information at any time point and at any image location in a sequence of images is a key requirement in many situations and applications. The proposed method relies on so-called labeled affine flows (LAF) involving both affine velocity vectors and affine motion classes. At every pixel a motion class is inferred from the affine motion model selected in a set of candidate models estimated over a collection of windows. Then, the image is subdivided in blocks where motion class histograms weighted by the affine motion vector magnitudes are computed. They are compared blockwise to histograms of normal behaviours with a dedicated distance. More specifically, we introduce the local outlier factor (LOF) to detect anomalous blocks. LOF is a local flexible measure of the relative density of data points in a feature space, here the space of LAF histograms. By thresholding the LOF value, we can detect an anomalous motion pattern in any block at any time instant of the video sequence. The threshold value is automatically set in each block by means of statistical arguments. We report comparative experiments on several real video datasets, demonstrating that our method is highly competitive for the intricate task of detecting different types of anomalous motion in videos. Specifically, we obtain very competitive results on all the tested datasets: 99.2 \% AUC for UMN, 82.8 \% AUC for UCSD, and 95.73 \% accuracy for PETS 2009, at the frame level.
Highlights
Motion analysis, with all its possible branches, i.e., motion detection (Goyette et al, 2014), motion estimation (Fortun et al, 2015), motion segmentation (Zhang and Lu, 2001), and motion recognition (Cedras and Shah, 1995), is a key processing step for difficult tasks related to video analysis, such as activity recognition (Aggarwal and Ryoo, 2011; Vishwakarma and Agrawal, 2013; Li et al, 2015b)
We present several experiments to assess the performance of our anomalous motion detection method
This dataset is not truly intended to assess anomalous motion detection on its own, and the most performing methods are those exploiting both appearance and motion (Antic and Ommer, 2011; Li et al, 2014). This dataset is a popular one in crowd anomaly detection, so we believed that it was worth evaluating our method on the UCSD dataset as well, in order to show the usefulness of our anomalous motion detection method for this task, and by doing so its versatility
Summary
With all its possible branches, i.e., motion detection (Goyette et al, 2014), motion estimation (Fortun et al, 2015), motion segmentation (Zhang and Lu, 2001), and motion recognition (Cedras and Shah, 1995), is a key processing step for difficult tasks related to video analysis, such as activity recognition (Aggarwal and Ryoo, 2011; Vishwakarma and Agrawal, 2013; Li et al, 2015b). Anomalous motion pertains to events of that type This kind of activity analysis usually requires intense human supervision, all the more when the objective of the analysis is identifying anomalies in the scene. A common setup for scenes where anomalies are sought consists of fixed-pose cameras pointing to scenes of interest. In these cases, the goal is to detect anomalies from the point of view of the camera. The goal is to detect anomalies from the point of view of the camera This task becomes even more difficult in crowded scenes, where the behavioral complexity in different parts of the video can cause confusion and distraction. The need for automatic systems that are able to assist the video monitoring of scenes has been growing steadily
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
