Detection Algorithm Research of Face Pose Based on Stereoscopic Vision and Biological Characteristics

Abstract:

The paper introduces a kind of detection method of face pose based on stereoscopic vision technology, approximately divides head’s deflexion into three plane rotations. By calculating the deflexion angle of three directions, you can determine the face’s pose. This method obtains face images by the left and right video channels, first analyses the similarity of double channels’ images to obtain three-dimensional information of face features’ key points. Then calculates three deflexion angles according to these information, therefore can correspondingly adjust and deform the original image to get standard frontal face image, and provides correction image for the latter face recognition. By this method the impact of pose change to face recognition can be reduced obviously in the earlier stage, so the system’s overall recognition accuracy rate is enhanced effectively.

• Periodical: Advanced Materials Research (Volume 460)
• Pages: 30-34
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.460.30
• Online since: February 2012
• Authors: Peng Xu, Yuan Men Zhou
• Keywords: Biological Characteristics, Face Recognition, Pose Detection, Stereoscopic Vision

References

[1] Lu Liu, Liwen Dong, Peng Jiang. Face Recognition Technology Research and Application[J]. Science and Technology Information(2010), p.492

[2] Xiuping Zhu, Xueyi Wu, Wenfeng Liu. Face Recognition and Prospect [J]. Computing and Information Technology(2008), pp.53-56

[3] Peng Xu, Yan Cai. Design of access control system based on stereo face recognition[J], CCASM 2010 - 2010 International Conference on Computer Application and System Modeling, Proceedings, Vol 14(2010), pp.14557-14561

DOI: 10.1109/iccasm.2010.5622171

[4] Kwang Ho An and Myung Jin Chung, 3D Head Tracking and Pose-Robust 2D Texture Map-Based Face Recognition using a Simple Ellipsoid Model[J], International Conference on Intelligent Robots and Systems(2008), pp.307-312

DOI: 10.1109/iros.2008.4650742

[5] Hu M, Worrall S, Sadka A H, et al. Face Feature Detection And Model Design For 2- D Scalable Model- Based Video Coding[M]. UK: VIE, Guildford-London(2003), pp.125-128

DOI: 10.1049/cp:20030503

[6] Hu M, Worrall S, Sadka A H, et al. A Fast and Efficient Chin Detection Method for 2- D Scalable Face Model Design[M]. UK: VIE, Guildford-London(2003), pp.121-124

[7] Mark. S. Nixon, Alberto S. Aguado. Feature Extraction and Image Processing, Second Edition[M]. Elsevier (Singapore) Pte Ltd(2008), pp.196-200