Weakly supervised video anomaly detection aims to locate abnormal activities in untrimmed videos without the need for frame-level supervision. Prior work has utilized graph convolution networks or self-attention mechanisms alongside multiple instance learning (MIL)-based classification loss to model temporal relations and learn discriminative features. However, these approaches are limited in two aspects: (1) Multi-branch parallel architectures, while capturing multi-scale temporal dependencies, inevitably lead to increased parameter and computational costs. (2) The binarized MIL constraint only ensures the interclass separability while neglecting the fine-grained discriminability within anomalous classes. To this end, we introduce a novel WS-VAD framework that focuses on efficient temporal modeling and anomaly innerclass discriminability. We first construct a Temporal Context Aggregation (TCA) module that simultaneously captures local-global dependencies by reusing an attention matrix along with adaptive context fusion. In addition, we propose a Prompt-Enhanced Learning (PEL) module that incorporates semantic priors using knowledge-based prompts to boost the discrimination of visual features while ensuring separability across anomaly subclasses. The proposed components have been validated through extensive experiments, which demonstrate superior performance on three challenging datasets, UCF-Crime, XD-Violence and ShanghaiTech, with fewer parameters and reduced computational effort. Notably, our method can significantly improve the detection accuracy for certain anomaly subclasses and reduced the false alarm rate. Our code is available at: https://github.com/yujiangpu20/PEL4VAD.

Automatic detection of violent videos has broad application prospects in many fields such as video surveillance and movie grading. However, most existing violent video detection models based on multimodal feature fusion ignore the fact that the audio-visual data in the same violent video may not semantically correspond. Blindly fusing non-corresponding features is not beneficial even potentially harmful to models. In this paper, we propose a novel violent video detection model based on semantic correspondence between audio-visual data from the same video. Deep neural networks are used to extract features of three different modalities: appearance, motion, and audio. After that, we choose the feature-level fusion strategy to fuse these multimodal features via shared subspace learning. Semantic correspondence is used to guide this process through multitask learning and semantic embedding learning. To evaluate t he effectiveness of our model, we conduct experiments on several public datasets and our self-built dataset: Violence Correspondence Detection. The results show that our model achieves quite competitive results on both.

In this paper, we propose and implement a novel method for recognition static hand gestures using depth data from Kinect sensor of Microsoft. Compared to the entire human body, the hand is a smaller object with more complex articulations and more easily affected by segmentation errors. So it is a very challenging problem to recognize hand gestures. Our approach involves choosing HOG feature with both geometric moment invariant features and adapted to the light transform by analyzing the features of hands characteristics. Through the rapid cascade Adaboost training algorithm obtains the training models of gestures and matches them, thus build the accuracy and efficiency hand gesture recognition system using the Kinect sensor.

We presented a static human gesture grading system based on the skeletal tracking module of Kinect sensor. We captured the 2-D skeleton wireframe of a poser's body and represented the data as limb vectors. The distance metric is defined as the included angle array computed between the real-time and the standard gesture. We proposed a grading formula to simulate the gesture judging scenario, of which the parameters can be adaptively computed or manually set. The system worked efficiently under the low computational complexity and was robust to input noise.

The aim of the violent recognition task is to determine whether a video contains violent behaviors. Given that violent behavior often comes with visual and audio anomalies, multimodal approaches have always played an important role in this field. However, existing methods have been limited by the insufficient utilization of audio-visual self-supervised semantic cues and correlation, resulting in a restricted representational capacity of the network and low generalization due to the scarcity of available violent video datasets. To address this issue, we propose a violent action recognition model based on global-local visual and audio contrastive learning. Our model introduces global and local contrastive objectives to achieve audio-visual multi-grained semantic alignment and leverage the correlation for violent video recognition. Experimental results demonstrate that our proposed model improves state-of-the-art by 2.31% on the VSD dataset, 0.71% on the Violent-Flows dataset, and 1.43% on the VCD dataset.