Title :
A novel 3D ear reconstruction method using a single image
Author :
Li, Chen ; Mu, Zhichun ; Zhang, Feng ; Wang, Shuai
Author_Institution :
Sch. of Autom. & Electr. Eng., Univ. of Sci. & Technol. Beijing, Beijing, China
Abstract :
To achieve denser 3D ear model from less controlled 2D image, we explore a 3D Ear Morphable Model (3DEMM) for 3D ear reconstruction using a single 2D ear image. Considering the unique structure of ear, we propose a novel dense corresponding method. The proposed method can overcome the shortcoming of optical flow based method and achieve pixel level dense correspondences based on physiological features of ear without choosing a reference ear. Novel 3D ear shape can be recovered from a single ear image based on the proposed 3D ear morphable model. Extensive experimental results have shown that our proposed method can obtain denser 3D ear model with lower cost and higher efficiency than existing methods.
Keywords :
ear; feature extraction; image reconstruction; image sequences; physiological models; solid modelling; stereo image processing; 3D ear morphable model; 3D ear reconstruction method; 3D ear shape recovery; 3DEMM; ear physiological feature; ear structure; optical flow based method; pixel level dense correspondence; single 2D ear image; Data models; Ear; Image reconstruction; Physiology; Shape; Solid modeling; Vectors; 3D Ear Morphable Model; 3D ear reconstruction using a single image; dense correspondence; sparse feature points;
Conference_Titel :
Intelligent Control and Automation (WCICA), 2012 10th World Congress on
Conference_Location :
Beijing
Print_ISBN :
978-1-4673-1397-1
DOI :
10.1109/WCICA.2012.6359404