Earticle

초록

영어

Anomaly recognition in visual and audio data has gained increasing significance in computer vision, as it plays a crucial role in protecting human lives and property. In this work, we developed a semi-supervised multimodal framework for anomaly recognition that combines audio and visual data for better performance. The proposed framework employs a hybrid network consisting of a convolutional neural network, Bi-Directional Long Short-Term Memory, a multi-head attention module, and a fully connected layer for anomalous pattern recognition. We created a novel real-time visual-audio anomaly recognition dataset and evaluated our framework on it, achieving promising results.

목차

요약
1. Introduction
2. Methodology
3. Experiment result
4. Conclusions
Acknowledgment
References

저자

• Amjid Ali [ Digital Contents Research Institute, Sejong University ]
• Hikmat Yar [ Digital Contents Research Institute, Sejong University ]
• Adnan Hussain [ Digital Contents Research Institute, Sejong University ]
• Altaf Hussain [ Digital Contents Research Institute, Sejong University ]
• Min Je Kim [ Digital Contents Research Institute, Sejong University ]
• Sung Wook Baik [ Digital Contents Research Institute, Sejong University ] Corresponding Author

참고문헌