Video-surveillance systems are becoming more and more autonomous in the detection and the reporting of abnormal events. In this context, this paper presents an approach to detect abnormal situations in crowded scenes by analyzing the motion aspect instead of tracking subjects one by one. The proposed approach estimates sudden changes and abnormal motion variations of a set of points of interest (POI). The number of tracked POIs is reduced using a mask that corresponds to hot areas of the built motion heat map. The approach detects events where local motion variation is important compared to previous events. Optical flow techniques are used to extract information such as density, direction and velocity. To demonstrate the interest of the approach, we present the results on the detection of collapsing events in real videos of airport escalator exits. 1.

This paper deals with the selection of relevant motion from multi-object movement. The proposed method is based on a multi-scale approach using features extracted from optical flow and global rarity quantification to compute bottom-up saliency maps. It shows good results from four objects to dense crowds with increasing performance. The results are convincing on synthetic videos, simple real video movements, a pedestrian database and they seem promising on very complex videos with dense crowds. This algorithm only uses motion features (direction and speed) but can be easily generalized to other dynamic or static features. Video surveillance, social signal processing and, in general, higher level scene understanding can benefit from this method. Index Terms — crowd analysis, social signal

Group action recognition in soccer videos is a challenging problem due to the difficulties of group action representation and camera motion estimation. This paper presents a novel approach for recognizing group action with a moving camera. In our approach, egomotion is estimated by the Kanade-Lucas-Tomasi feature sets on successive frames. The optical flow is then computed on compensated frames. Due to the inaccurate ego-motion estimation, the optical flow can not reflect accurate motion of objects. In this paper, we propose a new motion descriptor which treats the optical flow as spatial patterns and extracts accurate global motion from the noisy optical flow. The Latent-Dynamic Conditional Random Field model is employed to recognize group action. Experimental results show that our approach is promising. 1.

The use of algorithms which model part of the human attention can bring interesting results in the video analysis of difficult scenarios like dense crowds. A rarity-based attention model is able to provide in real-time areas in video frames where motion behavior is surprising compared to the rest of the motion in the same frame. This algorithm is also resistant to camera shake or translation and points out abnormal activities which can be used in surveillance but also to analyze and even foster social interaction. 1. VIDEO PROCESSING IN DENSE CROWDS Video processing for dense crowds is a field of computer vision which has specific properties. It is for example virtually impossible to obtain individual object tracking or it is difficult to acquire databases of specific events. If the

Abstract not found

Today’s rapid developments in digital media processing capabilities, and network speeds, make the dissemination of multimedia data extremely rapid and reliable, and have attracted significant research attention to video analysis, event detection, tracking and surveillance. In this work, a novel, generally applicable approach to the detection of human activity in video is presented. The areas of activity in the video are first detected via the accumulation and statistical processing of the motion vectors in all frames. The times (frames) at which events begin and end are defined as moments at which the statistical distribution of the motion vectors changes, for each pixel. These time instants are estimated in a novel manner, by applying sequential likelihood ratio testing on the motion vectors of the pixels that have been found to be active. The proposed system provides a theoretically sound solution for the detection of temporal changes in the human (or other) activity in video, without resorting to use of prior knowledge, heuristics, or ad-hoc thresholds. Sequential detection techniques allow us to find the frames where events begin and end, but also allows to pre-define the desired probabilities of false alarm and miss for the system. This is entirely novel for the temporal localization of activities and events in the video processing literature. Finally, sequential change detection methods require the smallest number of samples to detect a change, so they ensure the fastest detection of events. Experiments are performed with real sequences, involving human activities, for varying probabilities of false alarm and miss. Comparison with ground truth results shows that, indeed, the proposed method leads to meaningful localization of events both in time and in space.

top-down attention

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit:

journal homepage: www.elsevier.com/locate/cviu

The present work presents a new method for activity extraction and reporting from video based on the aggregation of fuzzy relations. Trajectory clustering is first employed mainly to discover the points of entry and exit of mobiles appearing in the scene. In a second step, proximity relations between resulting clusters of detected mobiles and contextual elements from the scene are modeled employing fuzzy relations. These can then be aggregated employing typical soft-computing algebra. A clustering algorithm based on the transitive closure calculation of the fuzzy relations allows building the structure of the scene and characterize the ongoing different activities of the scene. Discovered activity zones can be reported as activity maps with different granularities thanks to the analysis of the transitive closure matrix. Taking advantage of the soft relation properties, activity zones and related activities can be labeled in a more human-like language. We present results obtained on real videos corresponding to apron monitoring in the Toulouse airport in France. 1.