Future Frame Prediction Using Convolutional VRNN for Anomaly Detection
http://xxx.itp.ac.cn/pdf/1909.02168.pdf
https://github.com/yiweilu3/CONV-VRNN-for-Anomaly-Detection
一种新颖的基于变分自编码器 (VAE) 的序列生成模型,使用卷积LSTM (ConvLSTM) 来进行未来帧预测。
A novel sequential generative model based on variational
autoencoder (VAE) for future frame prediction with convolutional LSTM (ConvLSTM)
Object-centric Auto-encoders and Dummy Anomaliesfor Abnormal Event Detection in Video
https://arxiv.org/pdf/1812.04960.pdf
https://github.com/fjchange/object_centric_VAD
首先使用一种基于以对象为中心(Object-centric)的卷积自编码器(Convolutional
Auto-encoders)的无监督特征学习框架,对运动(motion)和外观(appearance)信息进行编码; 然后使用一种基于监督分类的方法对训练样本进行正态聚类分析。
First, we introduce an unsupervised feature learning framework based on object-centric convolutional auto-encoders to encode both motion and appearance information. Second, we propose a supervised classification approach based
on clustering the training samples into normality clusters.
Real-world Anomaly Detection in Surveillance Videos
https://arxiv.org/pdf/1801.04264.pdf
https://github.com/WaqasSultani/AnomalyDetectionCVPR2018
将正常和异常视频视为bags,将视频片段视为多实例学习(MIL)中的实例,并自动学习一个预测异常视频片段高异常分数的深度异常评测模型。此外,在损失函数中引入稀疏性和时间平滑性约束,以更好地定位训练过程中的异常。
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.
Future Frame Prediction for Anomaly Detection – A New Baseline
http://xxx.itp.ac.cn/pdf/1712.09867.pdf
https://github.com/StevenLiuWen/ano_pred_cvpr2018
第一个在视频预测任务中引入时间约束的工作。这种空间和运动约束有助于对正常事件进行未来帧预测,从而有助于识别不符合预期的异常事件。
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.