Research Article
Hardware Efficient Architecture with Variable Block Size for Motion Estimation
Table 5
Comparison among FPGA implementations of VBSME architectures.
| VBSME architecture | Search range | # of PEs | # of pixels | # of clock cycles to generate 41 SAD | # of clock cycles to generate MV | Frequency (MHz) | Frame processing rate (fps) | FPGA | LUTs |
| Olivares [12] | 32 × 32 | 256 | 16 | — | 4913 | 380.1 | 21.42 @1080p | Virtex 5 | 3768 | Elhamzi et al. [16] | 16 × 16 | 16 | 16 | — | 4096 | 436 | 13 @1080p | Virtex 6 | 1281 | Parandeh-Afshar et al. [15] | — | 4 | 4 | 64 | — | 285 | — | Virtex 2 | 1431 | Proposed | 32 × 32 | 17 | 16 | 17 | 272 | 393.16 | 179 @1080p | Virtex 5 | 9486 |
|
|