3D Conversion of 2D Video Encoded by H.264

Hong, Ho-Ki;Ko, Min-Soo;Seo, Young-Ho;Kim, Dong-Wook;Yoo, Ji-Sang

  • Received : 2011.09.02
  • Accepted : 2012.09.11
  • Published : 2012.11.01


In this paper, we propose an algorithm that creates three-dimensional (3D) stereoscopic video from two-dimensional (2D) video encoded by H.264 instead of using two cameras conventionally. Very accurate motion vectors are available in H.264 bit streams because of the availability of a variety of block sizes. 2D/3D conversion algorithm proposed in this paper can create left and right images by using extracted motion information. Image type of a given image is first determined from the extracted motion information and each image type gives a different conversion algorithm. The cut detection has also been performed in order to prevent overlapping of two totally different scenes for left and right images. We show an improved performance of the proposed algorithm through experimental results.


2D/3D conversion;Stereoscopic video;H.264;MTD;Delay factor;Motion vector


  1. R. Franich, R. Lagendijk and R. Horst, Reference model for hardware demonstrator implementation, RACE DISTIMA deliverable 45/TUD/IT/DS/B/003/bl, October, 1992.
  3. A. Rauol, State of the art of autostereoscopic displays, RACE DISTIMA deliverable 45/THO/WP4.2/DS/R/ 57/01, December 1995.
  4. Berlin, PANORAMA Final Demonstrations, AC092/ SIE/FinalDemo/DS/P/032 /b1, October 1998.
  5. S. Malassiotis, and M.G. Strintzis, "Coding of videoconference stereo image sequences using 3D models", Signal Processing: Image Communications 9 (1), pp. 125-135, January 1997.
  7. N. Hur and C. Ahn, "Experimental service of 3DTV broadcasting relay in Korea", Proc. SPIE 4864, pp.1- 13, 2002.
  8. C.V. Berkel, and D.W. Parker, "Multi-view 3D-LCD", Proc. SPIE 2653, pp. 32-39, 1996.
  9. R. Borner, "Autostereoscopic direct-view displays and rear-projection for short viewing distances by lenticular method", Proceedings of the First International Symposium on Three Dimensional Image Communication Technologies, pp.1-14, Tokyo, December 1993.
  10. Y. Matsumoto and et al., "Conversion system of monocular image sequence to stereo using motion parallax", SPIE Vol. 3012, Photonic West, 1997.
  11. T. Okino and et. al, "New television with 2D/3D image conversion techniques" SPIE Vol. 2653, Photonic West, 1995.
  12. Man Bae Kim, Sang Hoon Park "2D/3D video conversion based on key frame", HCI conference, Feb. 2002.
  15. M.T. Pourazad, P. Nasiopoulos, and R.K. Ward, "An H.264-based scheme for 2D to 3D video conversion," IEEE Trans. on Consumer Electronics, Vol. 55, No. 2, pp. 742-748, May, 2009.
  16. Feng Jie, Huang Hai, Zheng Yayu, and Zhu Wei, "Stereoscopic 3D conversion from 2D video in H.264/AVC compressed domain," 4th International Congress on Image and Signal Processing (CISP), Vol. 1, No .15-17, pp. 565-568, Oct. 2011.
  17. Ross, J., "Stereopsis by binocular delay", In Nature, Vol.248, pp.354-364, 1974.
  18. Ross, J. and Hogben, J. H., "Short-term memory in stereopsis", In Vision Research, Vol. 14, pp. 1195- 1201, 1974.
  19. ITU-R, "Methodology for the subjective assessment of the quality of television pictures," ITU-R Recommendation BT.500-11.

Cited by

  1. An investigation of pointing postures in a 3D stereoscopic environment vol.48, 2015,
  2. Stereo Matching with Confidence-Region Decomposition and Processing vol.14, pp.1, 2019,


Supported by : National Research Foundation of Korea (NRF)