Research Article
DetReco: Object-Text Detection and Recognition Based on Deep Neural Network
Table 3
Detection performance between the comprehensive dataset and modified comprehensive dataset.
| Detection frameworks | Dataset | mAP | Car | Bus | Person | Text |
| YOLOv3 544 × 544 | Comprehensive dataset | 76.9 | 89.3 | 87.9 | 83.5 | 66.0 | | Comprehensive dataset (modified) | 77.2 | 88.5 | 85.3 | 83.9 | 72.8 |
|
|