iphone 1 1 1 iphone,,., OpenGl ES 2.0 GLSL(OpenGL Shading Language), iphone GPGPU(General-Purpose Computing on Graphics Processing Unit)., AR Realtime Natural Feature Tracking Library for iphone Makoto Kondo, 1 Maki Sugimoto 1 and Masahiko Inami 1 We propose real time natural feature tracking library for ne gaming platform iphone. Using GLSL(OpenGL Shading Language) in OpenGL ES 2.0, this library does GPGPU(General-Purpose Computing on Grpahics Processing Unit) on iphone. This library enables users to make marker-less AR game application. 1.,, iphone Android. AppStore, Android Market,. 1 Graduate School of Media Design, Keio University, NintendoDS PSP,.,, OS,,., ios Android.,,,,.,, Augmented Reality AppStore Android Market 1)2).,,,., iphone3gs, iphone 4,. 2., Augmented Reality, Georg PTAM 3) iphone 4). PC, PTAM iphone,., Shi-Tomashi 5),., Wagner 6, SIFT 6), FAST 7)., CPU.,,,.. 3. iphone, iphone ipad OS, ios 4.0.,, ios. 1 c 2010 Information Processing Society of Japan
3.1 iphone. iphone 4,,, iphone3gs.. iphone 4 640x480 BGRA32, 420YpCbCr iphone 4 1280x720 BGRA32, 420YpCbCr iphone 3GS 640x480 BGRA32, 420YpCbCr 1 iphone Table 1 iphone Camera Specification,,, 640x480, OpenGL., iphone3gs, iphone4,., 420YpCbCr BGRA32, OpenGLES, BGRA32. 30fps,, 30fps,. 3.2 OpenGLES 2.0 3.2.1 iphone, 3GS, 4 OpenGLES 2.0 PoerVR SGX., iphone3g,,. iphone,.,,,,,,.,. 3.2.2 OpenGL, 8bit (GL UNSIGNED BYTE), (GL SHORT), (GL FLOAT),,,,, 0-255. GPU. 4. GPU, Sudipta N Sinha GPU- KLT 9). GPU-KLT,,,. ( 1 ), 4), 9) Shi-Tomasi 5). Shi-Tomasi, 2x2 Z ( ). ( ) Z = Ix2 IxIy I xi y I y 2 I x, I y x, y. ( 2 ),,,.,,.,,,,. 5. 5.1 GPU CPU GPU, glteximag2d,, GPU (1) 2 c 2010 Information Processing Society of Japan
CPU glreadpixels.,,,.,,.,,,.,,.,,,,.,,. 5.2,.. ( ) ( ) ( ) ( ) 1,, 1 a., iphone OpenGL ES 2.0 GLSL. ( 1 ),,., YUV, Y., BGR,, 1 Fig. 1 Image sequence.,. ( 2 ).,,,,.,,,., 3, 7, 1., 1.0. x, y, RGBA. 3 c 2010 Information Processing Society of Japan
, 0-255, 0.5(128).,. 1 b. ( 3 ), (1).,, I x 2, I xi y, I y 2. I x, I y R G, 0.5, 0.5, 0.5, R, G, B. 1 c b.. ( 4 ),. 7x7,,,. R, G, B, Ix2, IxIy, Iy2.,. 1 d c,. ( 5 ), (1). (1),. ( I 2 x + I 2 y )λ 2 + ( I 2 x I 2 2 y )λ + I x I 2 y ( I xi y) 2 = 0,, RGBA. 1 e d,. ( 6 ) CPU glreadpixels GPU., RGBA,.,,. ( 7 ),,.,,.,,.,,,. 5.3, CPU., 9) GPU KLT.,,,,,. 2 iphone,.,. 6. 3 iphone4,. 160px x 213px,.,, 160px x 213px, 10.05fps. 7., GPU, iphone,., GPU GPU.,, 4 c 2010 Information Processing Society of Japan
2 Fig. 2 Tracking image Fig. 3 3 Resolution and Processing Time., CPU., GPU, OpenGL ES 2.0 Android,. 1) Augmented Reality Broser:Layar http://.layar.com/ 2) http://sekaicamera.com/ja/ 3) G. Klein and D. Murray. Parallel tracking and mapping for small AR orkspaces. In Proc 6th IEEE and ACM International Symposium on Mixed and Augmented Reality (ISMAR 07), October 2007 4) Georg Klein and David Murray Parallel Tracking and Mapping on a Camera Phone In Proc. International Symposium on Mixed and Augmented Reality (ISMAR 09, Orlando) 5) Jianbo Shi and Carlo Tomasi. Good Features to Track. IEEE Conference on Computer Vision and Pattern Recognition, pages 593-600, 1994. 6) D. Loe, Distinctive Image Features from Scale-Invariant Key- points, Int l J. Computer Vision, vol. 60, no. 2, pp. 91-110, 2004. 7) E. Rosten and T. Drummond, Machine Learning for High-Speed Corner Detection, Proc. European Conf. Computer Vision (ECCV 06), pp. 430-443, 2006. 8) D. Wagner, G. Retimayr, A. Mulloni, T. Drummond, and D. Schmalstieg. Pose Tracking from natural features on mobile phones. In Proc 7th IEEE and ACM International Symposium on Mixed and Augmented Reality (ISMAR 08), Sept. 15-18 2008. 9) Sudipta N Sinha, Jan-Michael Frahm, Marc Pollefeys and Yakup Genc, GPU- Based Video Feature Tracking and Matching, Technical Report 06-012, Department of Computer Science, UNC Chapel Hill, May 2006. 5 c 2010 Information Processing Society of Japan