Results 1 -
8 of
8
Anomaly Detection and Localization in Crowded Scenes
"... Abstract—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 dynam ..."
Abstract
-
Cited by 5 (0 self)
- Add to MetaCart
(Show Context)
Abstract—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. Index Terms—Video analysis, surveillance, anomaly detection, crowded scene, dynamic texture, center-surround saliency Ç 1
The Role of Shape in Visual Recognition
"... Visual recognition requires a robust representation of typical object characteristics. Among all visual characteristics, shape plays a special role. It exhibits crucial invariance properties and captures the holistic structure of objects. However, shape cannot be extracted directly from an image, a ..."
Abstract
- Add to MetaCart
Visual recognition requires a robust representation of typical object characteristics. Among all visual characteristics, shape plays a special role. It exhibits crucial invariance properties and captures the holistic structure of objects. However, shape cannot be extracted directly from an image, as it is an emergent property. Thus, representing shape is challenging, since it is related to several key problems of computer vision, such as grouping, segmentation, and correspondence problems. This paper reviews the development of shape in object recognition so far, discusses the reasons for the underlying developmental trends, and presents some promising recent contributions that point towards more accurate models of object structure.
Text Mining and Image Anomaly Explanation with Machine Consciousness
"... In the present paper a series of implemented computer systems of information extraction for question answering that generate explanations of this extraction from text and image are presented. These computer generated explanations of the performance of a computer system may function as a criterion of ..."
Abstract
- Add to MetaCart
In the present paper a series of implemented computer systems of information extraction for question answering that generate explanations of this extraction from text and image are presented. These computer generated explanations of the performance of a computer system may function as a criterion of the exhibition of Machine Consciousness. Additionally the systems presented in the present paper may facilitate their communication with their users. Such communication is very useful for building the confidence of a user to a computer system. The main information extraction method used by the systems described here is the one based on a finite state automaton. Following the presentation of systems that have been implemented by our group for various text bases a novel development is reported concerning the explanation of image anomalies. The presentation of our novel work is helped by an example of anomaly detection and explanation in modern paintings.
BENSCH et al.: DEFORMABLE PROTOTYPES FOR MOTION ANOMALY DETECTION 1 Spatiotemporal Deformable Prototypes for Motion Anomaly Detection
"... This paper presents an approach for motion-based anomaly detection, where a proto-type pattern is detected and elastically registered against a test sample to detect anoma-lies in the test sample. The prototype model is learned from multiple sequences to define accepted variations. “Supertrajectorie ..."
Abstract
- Add to MetaCart
(Show Context)
This paper presents an approach for motion-based anomaly detection, where a proto-type pattern is detected and elastically registered against a test sample to detect anoma-lies in the test sample. The prototype model is learned from multiple sequences to define accepted variations. “Supertrajectories ” based on hierarchical clustering of dense point trajectories serve as an efficient and robust representation of motion patterns. An efficient hashing approach provides transformation hypotheses that are refined by a spatiotempo-ral elastic registration. We propose a new method for elastic registration of 3D+time trajectory patterns that induces spatial elasticity from trajectory affinities. The method is evaluated on a new motion anomaly dataset and performs well in detecting subtle anomalies. Moreover, we demonstrate the applicability to biological motion patterns. 1
Less is More: Video Trimming for Action Recognition
"... Action recognition is an important precursor for under-standing human activities in videos. The current paradigm of action recognition is to classify a video sequence as a whole. However, actions usually occur only in part of a video sequence, rendering the rest of the video irrelevant for action re ..."
Abstract
- Add to MetaCart
(Show Context)
Action recognition is an important precursor for under-standing human activities in videos. The current paradigm of action recognition is to classify a video sequence as a whole. However, actions usually occur only in part of a video sequence, rendering the rest of the video irrelevant for action recognition. In this paper, we propose a method for learning a subsequence classifier which can detect and classify part of a video that corresponds to the action. The subsequence classifier is trained from weakly labeled train-ing videos whose subsequence labels are not provided, but need to be inferred during learning. We use the frame-work of multiple instance learning to solve two problems jointly: i) find the action subsequences in training videos, ii) train the subsequence classifier using the inferred ac-tion subsequences. To obtain a robust solution to the MIL problem, we propose a sequential algorithm that consecu-tively decreases the number of inferred action subsequences per video and trims their length until only one short subse-quence is used as the action representative in each video. We evaluate the combination of the automatically trained subsequence classifier and the full sequence classifier on the very challenging Hollywood2 benchmark set and ob-serve a significant gain in the performance over the baseline full sequence classifier. Moreover, a favorable performance of the subsequence classifier for temporal localization of actions in videos is evidenced on two categories of the Hol-lywood2 dataset. 1.
NSH: Normality Sensitive Hashing for Anomaly Detection
"... Locality sensitive hashing (LSH) is a computationally efficient alternative to the distance based anomaly detec-tion. The main advantages of LSH lie in constant detection time, low memory requirement, and simple implementation. However, since the metric of distance in LSHs does not con-sider the pro ..."
Abstract
- Add to MetaCart
(Show Context)
Locality sensitive hashing (LSH) is a computationally efficient alternative to the distance based anomaly detec-tion. The main advantages of LSH lie in constant detection time, low memory requirement, and simple implementation. However, since the metric of distance in LSHs does not con-sider the property of normal training data, a naive use of existing LSHs would not perform well. In this paper, we propose a new hashing scheme so that hash functions are selected dependently on the properties of the normal train-ing data for reliable anomaly detection. The distance met-ric of the proposed method, called NSH (Normality Sensi-tive Hashing) is theoretically interpreted in terms of the re-gion of normal training data and its effectiveness is demon-strated through experiments on real-world data. Our results are favorably comparable to state-of-the arts with the low-level features. 1.
Research Article Anomaly Detection Based on Local Nearest Neighbor Distance Descriptor in Crowded Scenes
"... Copyright © 2014 Xing Hu et al.This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. We propose a novel local nearest neighbor distance (LN ..."
Abstract
- Add to MetaCart
(Show Context)
Copyright © 2014 Xing Hu et al.This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. We propose a novel local nearest neighbor distance (LNND) descriptor for anomaly detection in crowded scenes. Comparing with the commonly used low-level feature descriptors in previous works, LNND descriptor has two major advantages. First, LNND descriptor efficiently incorporates spatial and temporal contextual information around the video event that is important for detecting anomalous interaction among multiple events, while most existing feature descriptors only contain the information of single event. Second, LNND descriptor is a compact representation and its dimensionality is typically much lower than the low-level feature descriptor. Therefore, not only the computation time and storage requirement can be accordingly saved by using LNND descriptor for the anomaly detection method with offline training fashion, but also the negative aspects caused by using high-dimensional feature descriptor can be avoided. We validate the effectiveness of LNND descriptor by conducting extensive experiments on different benchmark datasets. Experimental results show the promising performance of LNND-based method against the state-of-the-artmethods. It is worthwhile to notice that the LNND-based approach requires less intermediate processing steps without any subsequent processing such as smoothing but achieves comparable event better performance. 1.
