Research Article
Triplet Deep Hashing with Joint Supervised Loss Based on Deep Neural Networks
Table 2
Number of triplets using the proposed triplet selection method.
| Dataset | Training dataset | Group | Category | Selected triplets | Time |
| CIFAR-10 | 5000 | 20 | 10 | 124,000 | 86 s | NUS-WIDE | 10,500 | 20 | 21 | 1,372,000 | 316 s | MS-COCO | 10,000 | 20 | 80 | 1,719,000 | 492 s |
|
|