Small Object Detection On Sod4Sb Public Test 1
評価指標
AP50
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | AP50 | Paper Title | Repository |
|---|---|---|---|
| Swin Transformer + Hierarchical design | 70.2 | Small Object Detection for Birds with Swin Transformer | - |
| GFL + Test Time Augmentation | 73.1 | BandRe: Rethinking Band-Pass Filters for Scale-Wise Object Detection Evaluation | |
| DL method (YOLOv8 + Ensamble) | 73.1 | MVA2023 Small Object Detection Challenge for Spotting Birds: Dataset, Methods, and Results | |
| Weighted Box Fusion (WBF) | 77.6 | Ensemble Fusion for Small Object Detection | - |
| E2 method (Normalized Gaussian Wasserstein Distance + Switch Hard Augmentation + Multi scale train + Weight Moving Average + CenterNet + VarifocalNet) | 69.6 | MVA2023 Small Object Detection Challenge for Spotting Birds: Dataset, Methods, and Results |
0 of 5 row(s) selected.