Kuroda, Kazuhiro 
  
3.2 Synthesis of Range Image And Video Image 
In general, range image obtained by 3D measurement system is digital image composed from discretely sampled points 
and so information between sampled points are lacked. We overcome this problem by texture mapping technique. 
In general, texture images are distorted by 3D shapes. But spatial resolution of our system is so high that above mentioned 
distortion is little. In this connection, Figures 4 show (a) measured shape data as it is, (b) approximated result by triangles 
and (c) rendered result by texture mapping technique. 
Figures 5 (a) show video image of existing human face, (b) front view of and (c) side view of measured range image 
rendered through the method in this paper. Figures 6 shows a series of 3D animation of speaking human face picked out 
from 32 scenes reconstructed from succesively measured range images. 
4 CONCLUSION 
In this paper, we tried to synthesize video images and successive range images obtained by high-speed and continuous 3D 
measurement system and to reconstruct realistic colored 3D surface model and 3D animation. 
In synthesizing, 3D points are mapped onto respective video images through simple coordinate system transformation and 
perspective transformation. Then we can obtain realistic 3D surface model and 3D animation through texture mapping 
technique briefly. In future, these method can be applied not only CG but also motion capture etc. 
(a)Video image (b)Front View (c)Side View 
  
Figures 5: Result of colored 3D image 
AKNOWLEGEMENT 
This work was partialy supported by Kayamori Foundation of Information Science Advancement and Foundation of 
Chukyo University for Advanced Research. 
REFERENCES 
Araki, K., Sato, M., Noda, T., Chiba, Y. and Shimizu, M., 1991. High speed and continuous rangefinding system. IEICE 
Trans. Special Issue on Computer Vision and Its Applications E74(10), pp. 3400-3406. 
Araki, K., Shimizu, M., Noda, T., Chiba, Y., Tsuda, Y., Ikegaya, K., Sannomiya, K. and Gomi, M., 1992. High speed and 
continuous 3-d measurement system. 11th IAPR International Conference on Pattern Recognition 4, pp. 62-65. 
Araki, K., Shimizu, M., Noda, T., Chiba, Y., Tsuda, Y., Ikegaya, K., Sannomiya, K. and Gomi, M., 1995. A high-speed 
and continuous 3d measurement system. Machine Vision and Applications 8(2), pp. 79-84. 
Fujita, K., Katayama, N., Kawai, T., Shimizu, M. and Araki, K., 1995. Segmentation of 3d object and animation of 
an existent human face using high-speed and continuous 3d measurement system. Trans. Institute of Image Electronics 
Engineers of Japan 24(5), pp. 567-575. 
  
470 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B5. Amsterdam 2000.