Table 1 Different benchmark datasets with proper training samples
From: DTCTH: a discriminative local pattern descriptor for image classification
Applications | Object classification | Event classification | Scene classification | Leaf classification | Facial expression classification | ||||
|---|---|---|---|---|---|---|---|---|---|
Databases | Caltech-256 | Caltech-101 | UIUC sports event | OT scene | Scene 15 | Indoor 67 | Swedish leaf | Cohn Kanade (CK) | CK+ |
Classes | 257 | 102 | 8 | 8 | 15 | 67 | 15 | 6/7 | 7 |
Total samples | 30,608 | 9145 | 1586 | 2688 | 4485 | 5620 | 1125 | 960/1280 | 981 |
Training images/class | 60 | 30 | 70 | 100 | 100 | 80 | 25 | Person independent | Â |
Test images/class | Rest | Rest | 60 | Rest | Rest | 20 | Rest | 10-fold cross-validation | Â |