Table 3 Model Comparison for various dataset
From: “Idol talks!” AI-driven image to text to speech: illustrated by an application to images of deities
References | Various object detection application area | Dataset | Accuracy |
|---|---|---|---|
Cao, Feng, et al. [34] | Remote sensing object detection (Dense Objects) | HRSC2016 and UCAS-AOD | 90.29% and 90.06% |
Chen, et al. [35] | UAV (Unmanned Aerial Vehicle) image-based vehicle detection | High-resolution UAV images | 79.5% to 91.9% |
Mahendrakar, Trupti, et al. [36] | Object detection for autonomous navigation | Not specified | Compared YOLOv5 and Faster R-CNN, performance metrics not explicitly mentioned |
Chen, Hao, et al. [37] | Road object detection | Custom road object dataset | mAP@0.5: 49.4% |
Horvat, Marko, and Gordan Gledec[38] | Image classification and localization | Not specified | Compared different YOLOv5 variations, performance metrics not provided |
Zhang, Jian, et al. [39] | Underwater object detection | URPC2019 and URPC2020 (underwater object datasets) | mAP@0.5: 79.8% (URPC2019) |
Proposed Model | Statue object detection | Deities dataset | 96% |
