Translator Disclaimer
29 April 2009 Thermal infrared exploitation for 3D face reconstruction
Author Affiliations +
Despite the advances in face recognition research, current face recognition systems are still not accurate or robust enough to be deployed in uncontrolled environments. The existence of a pose and illumination invariant face recognition system is still lacking. This research exploits the relationship between thermal infrared and visible imagery, to estimate 3D face with visible texture from infrared imagery. The relationship between visible and thermal infrared texture is learned using kernel canonical correlation analysis(KCCA), and then a 3D modeler is used to estimate the geometric structure from predicted visual imagery. This research will find it's application in uncontrolled environments where illumination and pose invariant identification or tracking is required at long range such as urban search and rescue (Amber alert, missing dementia patient), and manhunt scenarios.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Bernard O. Abayowa "Thermal infrared exploitation for 3D face reconstruction", Proc. SPIE 7347, Evolutionary and Bio-Inspired Computation: Theory and Applications III, 73470F (29 April 2009);

Back to Top