Search alternatives:
significant attention » significant potential (Expand Search), significant reduction (Expand Search)
increase decrease » increased release (Expand Search)
attention dataset » detection dataset (Expand Search), detection datasets (Expand Search), attention based (Expand Search)
significant attention » significant potential (Expand Search), significant reduction (Expand Search)
increase decrease » increased release (Expand Search)
attention dataset » detection dataset (Expand Search), detection datasets (Expand Search), attention based (Expand Search)
-
1
-
2
EMA attention mechanism working principle.
Published 2025“…We conducted ablation and comparative experiments on the public datasets RSOD and NWPU VHR-10. Compared to YOLOv8, our approach achieved improvements of 7.6% and 2.8% in mAP@0.5, and increases of 2.1% and 1.1% in mAP@0.5:0.95. …”
-
3
-
4
-
5
-
6
-
7
-
8
-
9
PR comparison on RSOD dataset.
Published 2025“…We conducted ablation and comparative experiments on the public datasets RSOD and NWPU VHR-10. Compared to YOLOv8, our approach achieved improvements of 7.6% and 2.8% in mAP@0.5, and increases of 2.1% and 1.1% in mAP@0.5:0.95. …”
-
10
Ablation study on the RSOD dataset.
Published 2025“…We conducted ablation and comparative experiments on the public datasets RSOD and NWPU VHR-10. Compared to YOLOv8, our approach achieved improvements of 7.6% and 2.8% in mAP@0.5, and increases of 2.1% and 1.1% in mAP@0.5:0.95. …”
-
11
-
12
Ablation study on the NWPU VHR-10 dataset.
Published 2025“…We conducted ablation and comparative experiments on the public datasets RSOD and NWPU VHR-10. Compared to YOLOv8, our approach achieved improvements of 7.6% and 2.8% in mAP@0.5, and increases of 2.1% and 1.1% in mAP@0.5:0.95. …”
-
13
PR comparison on NWPU VHR-10 dataset.
Published 2025“…We conducted ablation and comparative experiments on the public datasets RSOD and NWPU VHR-10. Compared to YOLOv8, our approach achieved improvements of 7.6% and 2.8% in mAP@0.5, and increases of 2.1% and 1.1% in mAP@0.5:0.95. …”
-
14
-
15
Comparison Results on CoNLL-2014 Dataset.
Published 2025“…The average edit distance decreases from 3.2 to 2.5, the BLEU score increases from 0.65 to 0.72, and the average processing time is reduced from 2.3 seconds to 1.8 seconds. …”
-
16
-
17
-
18
-
19
-
20