Table 4 Comparison of homologous rubbing alignment methods on time efficiency
From: An open benchmark for oracle bone rubbing image retrieval
Category | Method | Match Points | Feature Extract | Feature Match | Total Time |
|---|---|---|---|---|---|
End-to-end | MeCoq12 | N/A | N/A | N/A | 38s |
| Â | HiHPq43 | N/A | N/A | N/A | 72s |
| Â | Dino44 | N/A | N/A | N/A | 57s |
Feature-based | SIFT45 | 4,927 | 34ms (CPU) | 15.805ms (CPU) | 19.77h |
| Â | SuperGlue25 | 850 | 35ms (GPU) | 27.449ms (GPU) | 34.32h |
| Â | LightGlue47 | 767 | 42ms (GPU) | 18.653ms (GPU) | 23.34h |
| Â | DeDoDe26 | 10,000 | 172ms (GPU) | 13.921ms (GPU) | 17.53h |
| Â | SGMnet46 | 998 | 11ms (GPU) | 8.231ms (CPU) | 10.29h |
Content-based | MSHRR (ours) | 16 | 2263ms (GPU) | 0.200ms (CPU) | 2.13h |
