A method of face detection with deep models for patrol videos

Baozhi Jia; Ming Zhu

doi:10.1117/12.2197009

Translator Disclaimer

6 July 2015 A method of face detection with deep models for patrol videos

Baozhi Jia; Ming Zhu

Author Affiliations +

Proceedings Volume 9631, Seventh International Conference on Digital Image Processing (ICDIP 2015); 96311B (2015) https://doi.org/10.1117/12.2197009
Event: Seventh International Conference on Digital Image Processing (ICDIP15), 2015, Los Angeles, United States

Abstract

Common face detection methods may fail in videos captured by patrol cars for the low resolution and uncooperative situation. We proposed a method to handle this problem with a parts-based deep model. Different parts of human bodies are detected for improving the accuracy of face detection in this method. A deep neural network is used for combining the detections of different parts. Experiments were conducted on two different datasets. The results demonstrate that the proposed method outperforms existing common face detection methods.

Citation Download Citation

Baozhi Jia and Ming Zhu "A method of face detection with deep models for patrol videos", Proc. SPIE 9631, Seventh International Conference on Digital Image Processing (ICDIP 2015), 96311B (6 July 2015); https://doi.org/10.1117/12.2197009