Name
Abstract | ||
|---|---|---|
H.264 adopts new coding tools such as variable block size, quarter-pixel-accuracy motion estimation/compensation, multiple reference frames, loop filter, etc. The adoption of these tools enables a macroblock to have more information compared with previous standards. In H.264 each macroblock can have up to sixteen motion vectors, four reference frames, and a mode of the macroblock, all of which can be used for temporal error concealment. The H.264 error concealment in the informative part, however, uses a similar method to prior coding standards that consider only motion vectors of macroblocks adjacent to the lost macroblock. In this paper, we propose an effective temporal error concealment algorithm for H.264-coded video. It uses not only motion vectors and reference frames but the modes of macroblocks adjacent to the lost macroblock as well. Depending on the modes of neighboring macroblocks, each lost macroblock is concealed on the basis of different block sizes. Simulation results show the proposed method yields better video quality than conventional approaches. |
Year | DOI | Venue |
|---|---|---|
2005 | 10.1109/ICIP.2005.1530545 | 2005 International Conference on Image Processing (ICIP), Vols 1-5 |
Keywords | Field | DocType |
temporal error concealment, H.264 video coding, variable block size, macroblock mode | Reference frame,Macroblock,Computer vision,Block-matching algorithm,Quarter-pixel motion,Coding tree unit,Computer science,Motion compensation,Artificial intelligence,Motion estimation,Deblocking filter | Conference |
ISSN | Citations | PageRank |
1522-4880 | 13 | 0.91 |
References | Authors | |
5 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Dong-Hyung Kim | 1 | 130 | 14.38 |
| Siyoung Yang | 2 | 48 | 4.18 |
| Jechang Jeong | 3 | 1002 | 141.22 |