Activity Recognition on OmniFall
89.4CGSLogi-PAR
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Logi-PARCategory=Neuro-Sym, Backbone=Swin-B2026.03 | 89.4 | 91.5 | 93.1 | 88.2 | 90.5 | 91 | |
| Dual-CausalCategory=Causal, Backbone=ResNet-1012026.03 | 82.5 | 80.2 | 89.3 | 78.4 | 86.2 | 88.4 | |
| NS-VQA (Med)Category=Neuro-Sym, Backbone=ViT2026.03 | 79.4 | 75.1 | 83.2 | 72.1 | 84.5 | 85.8 | |
| ContextGPTCategory=Neuro-Sym, Backbone=LLMs2026.03 | 76.8 | 72.4 | 86.5 | 68.5 | 82.1 | 86.2 | |
| Med-flamingo (Few-Shot)Category=VLM, Backbone=Openflamingo-9b2026.03 | 74.1 | 61.2 | 79.8 | 45 | 60.1 | 75.4 | |
| VideoLLaMA2Category=VLM, Backbone=Mistral-7B2026.03 | 72.8 | 63.5 | 88.4 | 41.6 | 62.4 | 80.1 | |
| LanguageBindCategory=VLM, Backbone=CLIP-ViT-L2026.03 | 70.5 | 64.1 | 90.1 | 38.2 | 66.2 | 85.1 | |
| InternVideo2Category=VLM, Backbone=InternViT-6B2026.03 | 68.3 | 62.5 | 91.5 | 35.4 | 65.8 | 84.8 | |
| UniFormerV2Category=Vis-T, Backbone=UniFormer-L2026.03 | 64.5 | 60.1 | 90.1 | 14.2 | 78.4 | 83.5 | |
| VideoMAE-V2Category=Vis-T, Backbone=ViT-g2026.03 | 62.1 | 58.4 | 89.2 | 12.5 | 75.2 | 81 | |
| Swin-UNETRCategory=Medical-T, Backbone=Swin-B2026.03 | 58.4 | 55.2 | 87 | 10.1 | 80.5 | 80.2 | |
| ClinActNetCategory=Multimodal, Backbone=Transformer2026.03 | 58.2 | 53.1 | 86.1 | 11.2 | 73.1 | 78.4 | |
| ActionCLIPCategory=Vis-VLM, Backbone=ViT-B/162026.03 | 55.4 | 51.2 | 84.2 | 10.5 | 71.5 | 75.8 | |
| SlowFastCategory=Vis-CNN, Backbone=ResNet-502026.03 | 52.1 | 48.3 | 81.5 | 8.4 | 68.2 | 72.1 |