作者: Matthew Shreve , Vasant Manohar , Dmitry Goldgof , Sudeep Sarkar
DOI: 10.1109/BTAS.2010.5634524
关键词: Face (geometry) 、 Facial recognition system 、 Artificial intelligence 、 Optical flow 、 Adverse conditions 、 Principal component analysis 、 Shadow 、 Camouflage 、 Computer science 、 Open mouth 、 Computer vision
摘要: This paper presents a method for face identification under adverse conditions by combining regular, frontal images with facial strain maps using score-level fusion. Strain are generated calculating the central difference of optical flow field obtained from each subject's during open mouth expression. Subjects were recorded and without camouflage three lighting conditions: normal lighting, low strong shadow. Experimental results demonstrate that useful supplemental biometrie in all conditions, especially condition, where 30% increase rank 1 recognition is observed over baseline PCA-based algorithm.