Research Article
Deep Hashing Based Fusing Index Method for Large-Scale Image Retrieval
Table 3
Accuracy in terms of MAP compared to nondeep methods with deep features.
| Method | CIFAR-10 | NUS-WIDE | 12 bits | 24 bits | 32 bits | 48 bits | 12 bits | 24 bits | 32 bits | 48 bits |
| SH + CNN | 0.183 | 0.164 | 0.161 | 0.161 | 0.621 | 0.616 | 0.615 | 0.612 | ITQ + CNN | 0.237 | 0.246 | 0.255 | 0.261 | 0.719 | 0.739 | 0.747 | 0.756 | SPLH + CNN | 0.299 | 0.330 | 0.335 | 0.330 | 0.753 | 0.775 | 0.783 | 0.786 | LFH + CNN | 0.208 | 0.242 | 0.266 | 0.339 | 0.695 | 0.734 | 0.739 | 0.759 | KSH + CNN | 0.488 | 0.539 | 0.548 | 0.563 | 0.768 | 0.786 | 0.790 | 0.799 | SDH + CNN | 0.478 | 0.557 | 0.584 | 0.592 | 0.780 | 0.804 | 0.815 | 0.824 | FastH + CNN | 0.553 | 0.607 | 0.619 | 0.636 | 0.779 | 0.807 | 0.816 | 0.825 | DHFI | 0.613 | 0.750 | 0.768 | 0.774 | 0.807 | 0.836 | 0.854 | 0.860 |
|
|