Name
Abstract | ||
|---|---|---|
Motion compensation algorithm have not addressed properly the issue of motion detection. Also no separation is made between motion and noise. In this paper we separate noise from motion, we also define a norm of the degree of motion, we estimate the motion and we provide a motion vector to compensate for the motion. A comparison of the motion compensation algorithm which is a refinement of our previous work, with some of the most popular algorithms is given. The motion vector provides better than half a pixel accuracy. Our approach is capable of separating the motion from noise, provides better than half a pixel accuracy, and it takes less computations than such popular block matching algorithms as logarithmic search, conjugate direction search, three step search, and hierarchical search. |
Year | Venue | Keywords |
|---|---|---|
2000 | INTELLIGENT SYSTEMS | motion estimation |
Field | DocType | Citations |
Computer vision,Quarter-pixel motion,Motion detection,Computer science,Motion compensation,Artificial intelligence,Motion estimation | Conference | 0 |
PageRank | References | Authors |
0.34 | 1 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Evangelos A. Yfantis | 1 | 16 | 10.80 |
| A. Angelopoulos | 2 | 0 | 0.34 |
| A. Popovich | 3 | 0 | 0.34 |