Table 9 cpp and execution time for ARM Cortex-A9 and Cortex-A15
From: Covariance tracking: architecture optimizations for embedded systems
Sequence | Panda | Pedxing | ||
|---|---|---|---|---|
Size | 312×233 | 640×480 | ||
ARM Cortex-A9 | ||||
Algorithm version | SoA | AoS++ | SoA | AoS++ |
Features computation (cpp) | 461 | 461 | 486 | 486 |
Kernel computation (cpp) | 1491 | 395 | 1600 | 415 |
Tracking (cpp) | 96 | 96 | 19 | 19 |
Total (cpp) | 2048 | 952 | 2106 | 921 |
Kernel/total | 73% | 42% | 73% | 45% |
Total speedup | ×2.2 | ×2.3 | ||
1-C execution time (ms) | 149 | 69 | 647 | 283 |
2-C execution time (ms) | 108 | 36 | 492 | 149 |
ARM Cortex-A15 | ||||
Algorithm version | SoA | AoS++ | SoA | AoS++ |
Features computation (cpp) | 207 | 207 | 205 | 205 |
Kernel computation (cpp) | 562 | 170 | 582 | 177 |
Tracking (cpp) | 28 | 52 | 4 | 7 |
Total (cpp) | 797 | 429 | 791 | 389 |
Kernel/total | 70% | 39% | 73% | 45% |
Total speedup | ×1.9 | ×2.0 | ||
1-C execution time (ms) | 38 | 20 | 161 | 79 |
2-C execution time (ms) | 27 | 10 | 119 | 42 |