Creating 3D realistic head: from two orthogonal photos to multiview face contents

Abstract

3D Head models have many applications, such as virtual conference, 3D web game, and so on. The existing several webbased face modeling solutions that can create a 3D face model from one or two user uploaded face images, are limited to generating the 3D model of only face region. The accuracy of such reconstruction is very limited for side views, as well as hair regions. The goal of our research is to develop a framework for reconstructing the realistic 3D human head based on two approximate orthogonal views. Our framework takes two images, and goes through segmentation, feature points detection, 3D bald head reconstruction, 3D hair reconstruction and texture mapping to create a 3D head model. The main contribution of the paper is that the processing steps are applies to both the face region as well as the hair region.

Rapid 3D Face Modeling from Video

“…Because it is a very important step that profile facial feature points are detected accurately for the subsequent steps. But Snake model [16] is very sensitive to initial position and LMCT works well only for Mongoloid looking people [5]. Our method is used to avoid these limitations, which only needs profile facial silhouette and frontal facial points (see Fig.…”

“…In this paper, Active Shape Model (ASM) is used to extract 68 facial feature points (eyes, nose, mouth, and facial silhouette, etc.) on the frontal image [5]. The training data sets are provided by Song [8].…”

“…Lin et al also proposed a modeling method based on orthogonal images [5]. The feature points on the profile are detected automatically by using LMCT (Local Maximum-Curvature Tracking).…”