NWPU
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Classification | NWPU (test) | OA0.9506 | 80 | |
| Few-Shot Classification | NWPU | Accuracy93.25 | 54 | |
| Scene Classification | NWPU 20% training ratio 45 classes (test) | Overall Accuracy95.82 | 45 | |
| Text-to-Image Retrieval | NWPU (test) | R@113.62 | 44 | |
| Image-to-Text Retrieval | NWPU (test) | Recall@1 (R@1)22.84 | 44 | |
| Scene Classification | NWPU | Top-1 Acc96.92 | 38 | |
| Scene Classification | NWPU RESISC45 (test) | Top-1 Accuracy96.74 | 28 | |
| Object Detection | NWPU 2024 (test) | AP (bbox)76.04 | 22 | |
| Multimodal Retrieval-Augmented Generation | NWPU (test) | Accuracy96.74 | 22 | |
| Object Detection | NWPU | AP70.3 | 21 | |
| Scene Classification | NWPU 10/90 split | Accuracy94.56 | 21 | |
| Crowd Counting | NWPU (test) | MAE46.23 | 15 | |
| Crowd Counting | NWPU 49 | MAE71.7 | 13 | |
| Object Detection | NWPU | mAP5093.3 | 12 | |
| Object Detection | NWPU-10 (test) | Recall @ IoU=0.591.7 | 10 | |
| Crowd Counting | NWPU | MAE46.23 | 9 | |
| Open-Vocabulary Object Detection | NWPU (test) | mAP (PL)29.39 | 8 | |
| Crowd Counting | NWPU (val) | MAE70.5 | 8 | |
| Video Anomaly Detection | NWPU 43 | Frame-level AUC62.1 | 7 | |
| Image Segmentation | NWPU | mIoU84 | 6 | |
| Question Generation | NWPU-300 (test) | BLEU-141.87 | 5 | |
| Dataset Distillation | NWPU | Training Time (h)13.17 | 5 | |
| Crowd localization and counting | NWPU | F1 Score61.5 | 4 | |
| Crowd localization and counting | NWPU (test) | F1 Score51.3 | 4 | |
| Instance Segmentation | NWPU | IoU50.23 | 4 |
Showing 25 of 28 rows