A novel framework for anomaly detection in crowded scenes is presented. Three properties are identified as important for the design of a localized video representation suitable for anomaly detection in such scenes: 1) joint modeling of appearance and dynamics of the scene, and the abilities to detect 2) temporal, and 3) spatial abnormalities. The model for normal crowd behavior is based on mixtures of dynamic textures and outliers under this model are labeled as anomalies. Temporal anomalies are equated to events of low-probability, while spatial anomalies are handled using discriminant saliency. An experimental evaluation is conducted with a new dataset of crowded scenes, composed of 100 video sequences and five well defined abnormality categories. The proposed representation is shown to outperform various state of the art anomaly detection techniques.

/pdf/anomaly-detection-in-crowded-scenes-2ncde4zo7t.pdf

Anomaly detection in crowded scenes

Surveillance videos are able to capture a variety of realistic anomalies. In this paper, we propose to learn anomalies by exploiting both normal and anomalous videos. To avoid annotating the anomalous segments or clips in training videos, which is very time consuming, we propose to learn anomaly through the deep multiple instance ranking framework by leveraging weakly labeled training videos, i.e. the training labels (anomalous or normal) are at video-level instead of clip-level. In our approach, we consider normal and anomalous videos as bags and video segments as instances in multiple instance learning (MIL), and automatically learn a deep anomaly ranking model that predicts high anomaly scores for anomalous video segments. Furthermore, we introduce sparsity and temporal smoothness constraints in the ranking loss function to better localize anomaly during training. We also introduce a new large-scale first of its kind dataset of 128 hours of videos. It consists of 1900 long and untrimmed real-world surveillance videos, with 13 realistic anomalies such as fighting, road accident, burglary, robbery, etc. as well as normal activities. This dataset can be used for two tasks. First, general anomaly detection considering all anomalies in one group and all normal activities in another group. Second, for recognizing each of 13 anomalous activities. Our experimental results show that our MIL method for anomaly detection achieves significant improvement on anomaly detection performance as compared to the state-of-the-art approaches. We provide the results of several recent deep learning baselines on anomalous activity recognition. The low recognition performance of these baselines reveals that our dataset is very challenging and opens more opportunities for future work. The dataset is available at: http://crcv.ucf.edu/projects/real-world/

/pdf/real-world-anomaly-detection-in-surveillance-videos-4jb8p2a0v1.pdf

Real-World Anomaly Detection in Surveillance Videos

Speedy abnormal event detection meets the growing demand to process an enormous number of surveillance videos. Based on inherent redundancy of video structures, we propose an efficient sparse combination learning framework. It achieves decent performance in the detection phase without compromising result quality. The short running time is guaranteed because the new method effectively turns the original complicated problem to one in which only a few costless small-scale least square optimization steps are involved. Our method reaches high detection rates on benchmark datasets at a speed of 140-150 frames per second on average when computing on an ordinary desktop PC using MATLAB.

/pdf/abnormal-event-detection-at-150-fps-in-matlab-2hd6b50jp8.pdf

Abnormal Event Detection at 150 FPS in MATLAB

Anomaly detection in videos refers to the identification of events that do not conform to expected behavior. However, almost all existing methods tackle the problem by minimizing the reconstruction errors of training data, which cannot guarantee a larger reconstruction error for an abnormal event. In this paper, we propose to tackle the anomaly detection problem within a video prediction framework. To the best of our knowledge, this is the first work that leverages the difference between a predicted future frame and its ground truth to detect an abnormal event. To predict a future frame with higher quality for normal events, other than the commonly used appearance (spatial) constraints on intensity and gradient, we also introduce a motion (temporal) constraint in video prediction by enforcing the optical flow between predicted frames and ground truth frames to be consistent, and this is the first work that introduces a temporal constraint into the video prediction task. Such spatial and motion constraints facilitate the future frame prediction for normal events, and consequently facilitate to identify those abnormal events that do not conform the expectation. Extensive experiments on both a toy dataset and some publicly available datasets validate the effectiveness of our method in terms of robustness to the uncertainty in normal events and the sensitivity to abnormal events. All codes are released in https://github.com/StevenLiuWen/ano_pred_cvpr2018.

/pdf/future-frame-prediction-for-anomaly-detection-a-new-baseline-4n3l3cjdjp.pdf

Future Frame Prediction for Anomaly Detection - A New Baseline

The detection and localization of anomalous behaviors in crowded scenes is considered, and a joint detector of temporal and spatial anomalies is proposed. The proposed detector is based on a video representation that accounts for both appearance and dynamics, using a set of mixture of dynamic textures models. These models are used to implement 1) a center-surround discriminant saliency detector that produces spatial saliency scores, and 2) a model of normal behavior that is learned from training data and produces temporal saliency scores. Spatial and temporal anomaly maps are then defined at multiple spatial scales, by considering the scores of these operators at progressively larger regions of support. The multiscale scores act as potentials of a conditional random field that guarantees global consistency of the anomaly judgments. A data set of densely crowded pedestrian walkways is introduced and used to evaluate the proposed anomaly detector. Experiments on this and other data sets show that the latter achieves state-of-the-art anomaly detection results.

/pdf/anomaly-detection-and-localization-in-crowded-scenes-qkc6y7r8q4.pdf

Anomaly Detection and Localization in Crowded Scenes

We present a novel algorithm for detection of certain types of unusual events. The algorithm is based on multiple local monitors which collect low-level statistics. Each local monitor produces an alert if its current measurement is unusual and these alerts are integrated to a final decision regarding the existence of an unusual event. Our algorithm satisfies a set of requirements that are critical for successful deployment of any large-scale surveillance system. In particular, it requires a minimal setup (taking only a few minutes) and is fully automatic afterwards. Since it is not based on objects' tracks, it is robust and works well in crowded scenes where tracking-based algorithms are likely to fail. The algorithm is effective as soon as sufficient low-level observations representing the routine activity have been collected, which usually happens after a few minutes. Our algorithm runs in real-time. It was tested on a variety of real-life crowded scenes. A ground-truth was extracted for these scenes, with respect to which detection and false-alarm rates are reported.

Robust Real-Time Unusual Event Detection using Multiple Fixed-Location Monitors

A method of monitoring a target, which includes tracking the target using a tracking modality, thereby obtaining an estimated track of the target, imaging the target using an imaging modality, thereby obtaining an image of the target, associating the estimated track with the image, and displaying at least one datum related to the image along with the estimated track.

Tracking and imaging data fusion

A classification system including a training device and one or more classification device for classifying one or more vectors other than training vectors. The training device is for selecting which training classification algorithms best classifies a set of training vectors, and for finding a set of values, of parameters of a generic classification algorithm, that enable the generic classification algorithm to substantially emulate the selected training classification algorithm.

Generic classification system

A system for surveillance of a region of interest (fig. 1). The system includes a number of monitors (12 of fig. 1), each monitor operative to acquire and to transmit periodically a description of a respective portion of the region of interest. The monitors (12 of fig. 1) are operative to transmit the descriptions in a first communication channel with insufficient bandwidth for all the monitors to transmit the descriptions simultaneously. The system also includes a control station (14 of fig. 1) for allocating respective portions of the channel to the monitors for the transmissions according to a number of rules. One or more of the rules includes feedback from the monitors that is separate from the descriptions (18 and 20 of fig. 1).

David Reinitz

Papers

Robust Real-Time Unusual Event Detection using Multiple Fixed-Location Monitors

Tracking and imaging data fusion

Generic classification system

Surveillance system and method