Corpus ID: 198229642

Motion-Aware Feature for Improved Video Anomaly Detection

@inproceedings{Zhu2019MotionAwareFF,
  title={Motion-Aware Feature for Improved Video Anomaly Detection},
  author={Yi Zhu and S. Newsam},
  booktitle={BMVC},
  year={2019}
}
Motivated by our observation that motion information is the key to good anomaly detection performance in video, we propose a temporal augmented network to learn a motion-aware feature. [...] Key Method Furthermore, we incorporate temporal context into the Multiple Instance Learning (MIL) ranking model by using an attention block. The learned attention weights can help to differentiate between anomalous and normal video segments better. With the proposed motion-aware feature and the temporal MIL ranking model…Expand
Global Information Guided Video Anomaly Detection
TLDR
An end-to-end GIG anomaly detection framework for anomaly detection using the video-level annotations (i.e., weak labels) to first mine the global pattern cues by leveraging the weak labels in a GIG module, and build a spatial reasoning module to measure the relevance between vectors in spatial domain with the global cue vectors. Expand
Weakly Supervised Video Anomaly Detection via Center-Guided Discriminative Learning
TLDR
An anomaly detection framework, called Anomaly Regression Net (ARNet), which only requires video-level labels in training stage is proposed, which yields a new state-of-the-art result for video anomaly detection on ShanghaiTech dataset. Expand
Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning
TLDR
A novel and theoretically sound method, named Robust Temporal Feature Magnitude learning (RTFM), which trains a feature magnitude learning function to effectively recognise the positive instances, substantially improving the robustness of the MIL approach to the negative instances from abnormal videos. Expand
Unsupervised Video Anomaly Detection via Flow-based Generative Modeling on Appearance and Motion Latent Features
TLDR
An implicit two-path AutoEncoder (ITAE) that exploits the structure of a SlowFast network and focuses on spatial and temporal information through appearance (slow) and motion (fast) encoders, respectively, which result outperforms the state-of-the-art methods. Expand
Temporal Pooling in Inflated 3DCNN for Weakly-supervised Video Anomaly Detection
TLDR
This paper proposes a method consisting of an ideal temporal pooling strategy in inflated 3DCNN feature map along with LSTM model for temporal dependency encoding for weakly-supervised anomaly detection task and is validated on a large scale video anomaly detection dataset, namely UCF-crime. Expand
Weakly-supervised Video Anomaly Detection with Contrastive Learning of Long and Short-range Temporal Features
TLDR
This paper proposes a novel method, named Multiscale Temporal Network trained with top-K Contrastive Multiple Instance Learning (MTN-KMIL), to address the problem of weaklysupervised video anomaly detection, in which given videolevel labels for training, the snippets containing abnormal events are identified. Expand
Motion-Attentive Network for Detecting Abnormal Situationsin Surveillance Video
TLDR
This work proposes a means of preserving motion patterns in abnormal situations through a network called MA-Net, which solves representation problems caused by a loss of sparse information and irregular patterns. Expand
Discriminative Clip Mining for Video Anomaly Detection
TLDR
By integrating the DACM and attentive MIL, one novel anomaly detection framework is proposed to learn more contrastive anomalous and normal patterns, and thus higher recognition performance can be achieved. Expand
An Efficient Anomaly Recognition Framework Using an Attention Residual LSTM in Surveillance Videos
TLDR
This work presents an efficient light-weight convolutional neural network (CNN)-based anomaly recognition framework that is functional in a surveillance environment with reduced time complexity and outperforms state-of-the-art models within complex surveillance environments. Expand
Abnormal Situation Detection on Surveillance Video Using Object Detection and Action Recognition
TLDR
A deep neural network model, called AT-Net, is proposed that automatically detects abnormal situations in the surveillance video, and an automatic video surveillance system developed based on this network model is introduced. Expand
...
1
2
3
4
...

References

SHOWING 1-10 OF 29 REFERENCES
Learning Deep Representations of Appearance and Motion for Anomalous Event Detection
TLDR
This work proposes Appearance and Motion DeepNet (AMDN) which utilizes deep neural networks to automatically learn feature representations, and introduces a novel double fusion framework, combining both the benefits of traditional early fusion and late fusion strategies. Expand
Dynamic video anomaly detection and localization using sparse denoising autoencoders
TLDR
Experimental analysis on two benchmark data sets show that the proposed dynamic anomaly detection and localization system outperforms the state-of-the-art models in terms of false positive rate, while also showing a significant reduction in computation time. Expand
Real-World Anomaly Detection in Surveillance Videos
TLDR
The experimental results show that the MIL method for anomaly detection achieves significant improvement on anomaly detection performance as compared to the state-of-the-art approaches, and the results of several recent deep learning baselines on anomalous activity recognition are provided. Expand
Video anomaly detection and localization using hierarchical feature representation and Gaussian process regression
TLDR
A hierarchical framework for detecting local and global anomalies via hierarchical feature representation and Gaussian process regression, which outperforms the main state-of-the-art methods on this topic and achieves at least 80% detection rates based on three challenging datasets. Expand
Learning Temporal Regularity in Video Sequences
TLDR
This work proposes two methods that are built upon the autoencoders for their ability to work with little to no supervision, and builds a fully convolutional feed-forward autoencoder to learn both the local features and the classifiers as an end-to-end learning framework. Expand
Anomaly Detection and Localization in Crowded Scenes
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 videoExpand
DeepVS: A Deep Learning Based Video Saliency Prediction Approach
TLDR
A large-scale eye-tracking database of videos (LEDOV), which includes 32 subjects’ fixations on 538 videos, is established and an object-to-motion convolutional neural network (OM-CNN) is developed to predict the intra-frame saliency for DeepVS, which is composed of the objectness and motion subnets. Expand
Temporal Segment Networks: Towards Good Practices for Deep Action Recognition
Deep convolutional networks have achieved great success for visual recognition in still images. However, for action recognition in videos, the advantage over traditional methods is not so evident.Expand
Abnormal Event Detection at 150 FPS in MATLAB
TLDR
An efficient sparse combination learning framework based on inherent redundancy of video structures achieves decent performance in the detection phase without compromising result quality and reaches high detection rates on benchmark datasets at a speed of 140-150 frames per second on average. Expand
Long-Term Temporal Convolutions for Action Recognition
TLDR
It is demonstrated that LTC-CNN models with increased temporal extents improve the accuracy of action recognition and the impact of different low-level representations, such as raw values of video pixels and optical flow vector fields, and the importance of high-quality optical flow estimation for learning accurate action models. Expand
...
1
2
3
...