Table 5 Results of ablation experiments conducted on the network architecture.
From: Multi-axis compression fusion network for vehicle re-identification
Method | mAP | Rank-1 | Rank-5 |
|---|---|---|---|
Baseline | 76.2% | 92.6% | 97.9% |
Baseline+S-MCF | 78.2% | 93.5% | 98.3% |
Baseline+C-MCF | 78.5% | 94.0% | 98.4% |
Bg+Bs (w/o RD) | 80.6% | 94.5% | 98.6% |
Bg+Bc (w/o RD) | 80.3% | 94.5% | 98.5% |
Bs+Bc (w/o RD) | 82.1% | 95.0% | 98.7% |
MCF-Net (w/o RD) | 83.1% | 95.5% | 98.8% |
MCF-Net (Ours) | 83.9% | 96.9% | 99.0% |
