Action Recognition on SSv2 Few-shot
66.7Top-1 Acc (5-way 1-shot)CLIP-SPM
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| CLIP-SPMPre-training=CLIP-ViT-B2025.12 | 66.7 | — | 77.3 | 74.8 | |
| CLIP-MEIPre-training=CLIP-ViT-B2025.12 | 66.3 | — | 79 | 75.9 | |
| DISTPre-training=CLIP-ViT-B2026.02 | 64.2 | — | 75.2 | — | |
| CLIP-SPMPre-training=CLIP-RN502025.12 | 62.3 | — | 69.4 | 67 | |
| CLIP-FSARModality=Multi-modal, Pre-training=CLIP-ViT-B, Tuning=FFT2024.11 | 61.9 | — | 72.1 | — | |
| CLIP-FSARPre-training=CLIP-ViT-B2025.12 | 61.9 | — | 72.1 | 68.1 | |
| CLIP-FSARPre-training=CLIP-ViT-B2026.02 | 61.9 | — | 72.1 | — | |
| CLIP-MEIPre-training=CLIP-RN502025.12 | 61.6 | — | 68.5 | 65.9 | |
| TAMTModality=Unimodal, Pre-training=ViT-B, Tuning=PEFT2024.11 | 61.4 | — | 73.3 | — | |
| MASTAFModality=Unimodal, Pre-training=JFT-ViT-B, Tuning=FFT2024.11 | 60.7 | — | — | — | |
| CLIP-CPM2CModality=Multi-modal, Pre-training=CLIP-ViT-B, Tuning=FFT2024.11 | 60.1 | — | 72.8 | — | |
| CLIP-M2DFPre-training=CLIP-ViT-B2025.12 | 60.1 | — | 72.7 | 68.9 | |
| CLIP-CPM2CPre-training=CLIP-ViT-B2025.12 | 60.1 | — | 72.8 | 68.9 | |
| CLIP-FSARPre-training=CLIP-RN502025.12 | 58.1 | — | 62.8 | 60.7 | |
| CLIP-FSARPre-training=CLIP-RN502026.02 | 58.1 | — | 62.8 | — | |
| CLIP-CPM2CPre-training=CLIP-RN502025.12 | 58 | — | 64 | 61.8 | |
| CLIP-M2DFPre-training=CLIP-RN502025.12 | 56.9 | — | 68 | 63.9 | |
| MoLoPre-training=INet-RN502025.12 | 56.6 | — | 70.6 | 67 | |
| MoLoPre-training=INet-RN502026.02 | 56.6 | — | 70.6 | — | |
| Lite-KMDPre-training=INet-RN502026.02 | 55.7 | — | 69.9 | — | |
| HyRSMPre-training=INet-RN502025.12 | 54.3 | — | 69 | 65.1 | |
| HyRSMPre-training=INet-RN502026.02 | 54.3 | — | 69 | — | |
| CapFSARModality=Multi-modal, Pre-training=BLIP-ViT-B, Tuning=FFT2024.11 | 54 | — | 70.1 | — | |
| HyRSMModality=Unimodal, Pre-training=BLIP-ViT-B(V), Tuning=FFT2024.11 | 52.1 | — | 69.5 | — | |
| CapFSARPre-training=BLIP-ViT-B2026.02 | 51.9 | — | 68.2 | — | |
| OTAMModality=Unimodal, Pre-training=CLIP-ViT-B(V), Tuning=FFT2024.11 | 50.2 | — | 68.6 | — | |
| CPMPre-training=INet-RN502026.02 | 49.3 | — | 66.7 | — | |
| TA2NPre-training=INet-RN502025.12 | 47.6 | — | 61 | — | |
| HCLPre-training=INet-RN502026.02 | 47.3 | — | 64.9 | — | |
| DPNPre-training=INet-RN502025.12 | 46.8 | — | 64.7 | — | |
| SloshNetPre-training=INet-RN502026.02 | 46.5 | — | 68.3 | — | |
| MTFANPre-training=INet-RN502025.12 | 45.7 | — | 60.4 | — | |
| MTFANPre-training=INet-RN502026.02 | 45.7 | — | 60.4 | — | |
| TRXModality=Unimodal, Pre-training=BLIP-ViT-B(V), Tuning=FFT2024.11 | 45.1 | — | 68.5 | — | |
| STRMPre-training=INet-RN502026.02 | 43.1 | — | 68.1 | — | |
| STRMPre-training=INet-RN502025.12 | 42.9 | — | 68.1 | — | |
| OTAMPre-training=INet-RN502025.12 | 42.8 | — | 52.3 | — | |
| OTAMPre-training=INet-RN502026.02 | 42.8 | — | 52.3 | — | |
| TRXPre-training=INet-RN502025.12 | 42 | — | 64.6 | — | |
| TRXPre-training=INet-RN502026.02 | 42 | — | 64.6 | — | |
| CLIPModality=Multi-modal, Pre-training=CLIP-ViT-B, Tuning=Frozen2024.11 | 30 | — | 42.4 | — | |
| CLIP-FreezePre-training=CLIP-ViT-B2026.02 | 30 | — | 42.4 | — | |
| A5Shots (K)=2, Text Augmentation=None2024.04 | — | 4.4 | — | — | |
| A5Shots (K)=4, Text Augmentation=None2024.04 | — | 5.1 | — | — | |
| A5Shots (K)=8, Text Augmentation=None2024.04 | — | 6.1 | — | — | |
| A5Shots (K)=16, Text Augmentation=None2024.04 | — | 9.7 | — | — | |
| ActionCLIPShots (K)=2, Text Augmentation=None2024.04 | — | 4.1 | — | — | |
| ActionCLIPShots (K)=4, Text Augmentation=None2024.04 | — | 5.8 | — | — | |
| ActionCLIPShots (K)=8, Text Augmentation=None2024.04 | — | 8.4 | — | — | |
| ActionCLIPShots (K)=16, Text Augmentation=None2024.04 | — | 11.1 | — | — | |
| HyRSMsetting=Any-shot2024.07 | — | 62.2 | — | — | |
| MoLosetting=Any-shot2024.07 | — | 63.7 | — | — | |
| OSTShots (K)=2, Text Augmentation=LLM-based2024.04 | — | 7 | — | — | |
| OSTShots (K)=4, Text Augmentation=LLM-based2024.04 | — | 7.7 | — | — | |
| OSTShots (K)=8, Text Augmentation=LLM-based2024.04 | — | 8.9 | — | — | |
| OSTShots (K)=16, Text Augmentation=LLM-based2024.04 | — | 12.2 | — | — | |
| SOAP-Netsetting=Any-shot2024.07 | — | 70.2 | — | — | |
| TC-CLIPShots (K)=2, Text Augmentation=None2024.04 | — | 7.3 | — | — | |
| TC-CLIPShots (K)=4, Text Augmentation=None2024.04 | — | 8.6 | — | — | |
| TC-CLIPShots (K)=8, Text Augmentation=None2024.04 | — | 9.3 | — | — | |
| TC-CLIPShots (K)=16, Text Augmentation=None2024.04 | — | 14 | — | — | |
| TC-CLIPShots (K)=2, Text Augmentation=LLM-based2024.04 | — | 7.3 | — | — | |
| TC-CLIPShots (K)=4, Text Augmentation=LLM-based2024.04 | — | 8.6 | — | — | |
| TC-CLIPShots (K)=8, Text Augmentation=LLM-based2024.04 | — | 9.3 | — | — | |
| TC-CLIPShots (K)=16, Text Augmentation=LLM-based2024.04 | — | 14 | — | — | |
| TRXsetting=Any-shot2024.07 | — | 61.3 | — | — | |
| Vanilla CLIPShots (K)=2, Text Augmentation=None2024.04 | — | 2.7 | — | — | |
| ViFi-CLIPShots (K)=2, Text Augmentation=None2024.04 | — | 6.2 | — | — | |
| ViFi-CLIPShots (K)=4, Text Augmentation=None2024.04 | — | 7.4 | — | — | |
| ViFi-CLIPShots (K)=8, Text Augmentation=None2024.04 | — | 8.5 | — | — | |
| ViFi-CLIPShots (K)=16, Text Augmentation=None2024.04 | — | 12.4 | — | — | |
| X-CLIPShots (K)=2, Text Augmentation=None2024.04 | — | 3.9 | — | — | |
| X-CLIPShots (K)=4, Text Augmentation=None2024.04 | — | 4.5 | — | — | |
| X-CLIPShots (K)=8, Text Augmentation=None2024.04 | — | 6.8 | — | — | |
| X-CLIPShots (K)=16, Text Augmentation=None2024.04 | — | 10 | — | — |