Image Classification on Oxford Flowers (Base, Novel, HM)
99Base AccuracyMMA
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MMABackbone=ViT-L/14, Shots=162025.03 | 99 | 80.2 | 88.61 | |
| MMABackbone=ViT-L/14, Shots=162025.03 | 99 | 80.2 | 88.61 | |
| 2SFSBackbone=ViT-L/14, Shots=162025.03 | 98.99 | 80.73 | 88.93 | |
| 2SFSBackbone=ViT-L/14, Shots=162025.03 | 98.99 | 80.73 | 88.93 | |
| 2SFSVisual Backbone=ViT-L/14, Shots (k)=82025.03 | 98.67 | 81.21 | 89.09 | |
| 2SFSBackbone=ViT-B/16, Shots=162025.03 | 98.29 | 76.17 | 85.83 | |
| SCPTShots=162026.07 | 98.08 | 75.89 | 85.57 | |
| ATPromptShots=162026.07 | 98.02 | 69.28 | 81.18 | |
| MMAVisual Backbone=ViT-L/14, Shots (k)=82025.03 | 97.97 | 80.3 | 88.26 | |
| CLIP-LoRABackbone=ViT-B/16, Shots=162025.03 | 97.91 | 68.61 | 80.68 | |
| MMABackbone=ViT-B/16, Shots=162025.03 | 97.77 | 75.93 | 85.48 | |
| TCPShots=162026.07 | 97.73 | 75.57 | 85.23 | |
| CoOpShots=162026.07 | 97.63 | 69.55 | 81.23 | |
| CoOpBackbone=ViT-B/16, Shots=162025.03 | 97.6 | 59.67 | 74.06 | |
| 2SFSBackbone=ViT-B/16, Shots=82025.03 | 96.68 | 76.36 | 85.33 | |
| 2SFSBackbone=ViT-B/16, Shots=8, Evaluation Protocol=base-to-novel generalization2025.03 | 96.68 | 76.36 | 85.33 | |
| 2SFSBackbone=ViT-B/32, Shots=162025.03 | 96.64 | 70.02 | 81.2 | |
| MaPLeBackbone=ViT-B/16, Shots=162025.03 | 95.92 | 72.46 | 82.56 | |
| 2SFSBackbone=ViT-B/32, Shots=82025.03 | 95.79 | 71.35 | 81.78 | |
| ProGradBackbone=ViT-B/16, Shots=162025.03 | 95.54 | 71.87 | 82.03 | |
| MMABackbone=ViT-B/32, Shots=162025.03 | 95.5 | 71.57 | 81.82 | |
| MMABackbone=ViT-B/16, Shots=82025.03 | 95.37 | 75.43 | 84.24 | |
| MMABackbone=ViT-B/16, Shots=8, Evaluation Protocol=base-to-novel generalization2025.03 | 95.37 | 75.43 | 84.24 | |
| KgCoOpBackbone=ViT-B/16, Shots=162025.03 | 95 | 74.73 | 83.65 | |
| KgCoOpShots=162026.07 | 95 | 74.73 | 83.65 | |
| 2SFSBackbone=ViT-B/16, Shots=42025.03 | 94.94 | 76.26 | 84.58 | |
| CoCoOpBackbone=ViT-B/16, Shots=162025.03 | 94.87 | 71.75 | 81.71 | |
| CoCoOpShots=162026.07 | 94.87 | 71.75 | 81.71 | |
| 2SFSVisual backbone=ViT-B/32, Shots per base class=4, Generalization setting=base-to-novel2025.03 | 93.99 | 71.7 | 81.35 | |
| 2SFSBackbone=ViT-B/32, Shots=42025.03 | 93.99 | 71.7 | 81.35 | |
| MMABackbone=ViT-B/32, Shots=82025.03 | 92.97 | 71.63 | 80.92 | |
| MMABackbone=ViT-B/16, Shots=42025.03 | 91.07 | 75.07 | 82.3 | |
| MMAVisual backbone=ViT-B/32, Shots per base class=4, Generalization setting=base-to-novel2025.03 | 87.03 | 70.83 | 78.1 | |
| MMABackbone=ViT-B/32, Shots=42025.03 | 87.03 | 70.83 | 78.1 | |
| CLIPBackbone=ViT-L/14, Shots=162025.03 | 80.34 | 83.05 | 81.67 | |
| CLIPVisual Backbone=ViT-L/14, Shots (k)=82025.03 | 80.34 | 83.05 | 81.67 | |
| CLIPBackbone=ViT-L/14, Shots=Zero-shot2025.03 | 80.34 | 83.05 | 81.67 | |
| ProTextShots=162026.07 | 74.36 | 78.44 | 76.35 | |
| CLIPBackbone=ViT-B/32, Shots=162025.03 | 72.36 | 73.69 | 73.02 | |
| CLIPVisual backbone=ViT-B/32, Shots per base class=4, Generalization setting=base-to-novel2025.03 | 72.36 | 73.69 | 73.02 | |
| CLIPBackbone=ViT-B/32, Shots=42025.03 | 72.36 | 73.69 | 73.02 | |
| CLIPBackbone=ViT-B/32, Shots=82025.03 | 72.36 | 73.69 | 73.02 | |
| CLIPBackbone=ViT-B/16, Shots=82025.03 | 72.08 | 77.8 | 74.83 | |
| CLIPBackbone=ViT-B/16, Shots=42025.03 | 72.08 | 77.8 | 74.83 | |
| CLIPBackbone=ViT-B/16, zero-shot=true2025.03 | 72.08 | 77.8 | 74.83 | |
| CLIPBackbone=ViT-B/16, Shots=8, Evaluation Protocol=base-to-novel generalization2025.03 | 72.08 | 77.8 | 74.83 | |
| CLIPShots=162026.07 | 72.08 | 77.8 | 74.83 |