Model-Based Multi-view Face Construction and Recognition in Videos
Model-based face construction and recognition in videos is a fundamental topic in image processing and video representation, while analysis faces across multiple views is more challenging than that from a fixed view because of the severe non-linearity caused by rotation in depth, self-occlusion, self-shading and illumination. In this paper, a novel method is presented to model and recognize multi-view faces in video sequences. Firstly, we design a multi-view face model to extract the face feature points. Secondly, a hybrid tracking method integrated optical flow with mean shift is proposed to estimate the face posture. Then, by using faces’ paths in different view and feature points obtained from models, a multi-view face map is synthesized by reconstruction and stitching the paths together. Finally, recognition experiments are conducted to evaluate the performance of our proposed approach.
KeywordsFace recognition Video-based face recognition Image stitching Active Appearance Model
Unable to display preview. Download preview PDF.
- 2.Lucas, B.D., Kanade, T.: An Iterative Image Registration Technique with An Application to Stereo vision. IJCAI 2, 674–679 (1981)Google Scholar
- 4.Hu, C., Harguess, J., Aggarwal, J.K.: Patch-based Face Recognition From Video. In: ICIP, pp. 3321–3324 (2009)Google Scholar
- 5.Ashraf, A.B., Lucey, S., Chen, T.: Learning Patch Correspondences for Improved Viewpoint Invariant Face Recognition. In: CVPR, pp. 1–8 (2008)Google Scholar
- 6.Viola, P., Jones, M.: Rapid Object Detection using A Boosted Cascade of Simple Features. In: CVPR, vol. 511 (2001)Google Scholar
- 7.Carnegie, R.C.: Mean-shift Blob Tracking through Scale Space. In: CVPR, vol. 2, pp. 234–240 (2003)Google Scholar
- 8.Uyttendaele, M., Eden, A., Skeliski, R.: Eliminating Ghosting And Exposure Artifacts in Image Mosaics. In: CVPR, vol. 2 (2001)Google Scholar