Image Captioning on NWPU RESISC45 Captions
88.9BLEU-1Ours
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| OursVisual Encoder=CLIP-Cap-4, Text Decoder=LlamaV22025.12 | 88.9 | — | 79.1 | 45.8 | 68.9 | 206.1 | 81.2 | 74.8 | 33.4 | |
| RS-CapRetVisual Encoder=CLIP-Cap-4, Text Decoder=LlamaV2, Finetuned=false2025.12 | 87.1 | — | 77.5 | 43.9 | 65 | 191.9 | 78.6 | 71.3 | 32 | |
| RS-CapRetVisual Encoder=CLIP-Cap-4, Text Decoder=LlamaV2, Finetuned=true2025.12 | 87.1 | — | 77.6 | 43.6 | 65.6 | 192.9 | 78.7 | 71.7 | 31.1 | |
| MLCA-NETVisual Encoder=VGG16, Text Decoder=LSTM2025.12 | 74.5 | — | 60.1 | 33.7 | 47.8 | 116.4 | 62.4 | 54.1 | 28.5 |