1 / 17

Adaptive Partial Distortion Search for Efficient Block Motion Estimation

Explore the Adaptive Partial Distortion Search (APDS) technique for speeding up the computation of accumulated partial distortion in block motion estimation. This method identifies representative pixels based on high pixel activities, edges, and texture to enhance accuracy. Learn how the Hilbert scan preserves edge information effectively compared to raster scan, improving motion estimation. Discover the APDS Algorithm and its efficiency in reducing computational operations.

tambre
Télécharger la présentation

Adaptive Partial Distortion Search for Efficient Block Motion Estimation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. BLOCK MOTION ESTIMATION USING ADAPTIVE PARTIALDISTORTION SEARCH Yui-Lam Chan, Wan-Chi Siu and Ko-Cheung Hui Centre for Multimedia Signal Processing Department of Electronic and Information Engineering The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong ICIP 2002

  2. Accumulated partial distortion

  3. Accumulated partial distortion

  4. Sub-sampling PDS (SSPDS)

  5. Speed up of the PDS • How fast the computation of the accumulated partial distortion is stopped. • The proposed technique • Based on the fact that high activities in pixels contribute most to the MAD criterion.

  6. Representative pixels • The representative pixels in the block • Pixel with high activities • Edges • Texture

  7. Hilbert scan and its features • Pass through all the grid points only once in a 2-D space. • 1891, German mathematician Hilbert • The edge info. In a 2-D image is preserved in its 1-D Hilbert-scan sequence more effectively than the rasterscan which may miss edges due to its scanning direction.

  8. 16x16 block Hilbert scan Raster scan Yankang Wang, Yanqun Wang and H. Kuroda, “A globally adaptive pixel-decimation algorithm for block-motion estimation, “ IEEE Trans. On Circuits and Syst. For Video Tech., Vol. 10 pp.1006-1011, Sept. 2000.

  9. The Hilbert scan

  10. The proposed APDS Algorithm

  11. The proposed APDS Algorithm

  12. An example of formatting a final sorted sequence, hs’

  13. An example of formatting a final sorted sequence, hs’

  14. The proposed APDS Algorithm

  15. Simulation Results 4.73 Average operations per block for different sequences. 3.76 A speed up about 1.96 to 4.73 3.45 1.96

  16. Simulation Results Frame-by-frame comparison of average operations per block in the “Salesman” sequence.

More Related