Temporal Action Localization on THUMOS14 v1.0 (75%-25%)
66.4mAP@0.3MGCA-Net
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| MGCA-NetVenue=-, Prompt Tuning=false, Text Feature=CLIP-B2025.11 | 66.4 | 59.7 | 49.6 | 38.2 | 26.2 | 48 | |
| Ti-FADVenue=NeurIPS’24, Prompt Tuning=false, Text Feature=CLIP-B2025.11 | 64 | 58.5 | 49.7 | 37.7 | 24.1 | 46.8 | |
| ZEETADTraining paradigm=training-based2026.05 | 61.4 | 53.9 | 44.7 | 34.5 | 20.5 | 43.2 | |
| ZEETADVenue=WACV’24, Prompt Tuning=true, Text Feature=CLIP-B2025.11 | 61.4 | 53.9 | 44.7 | 34.5 | 20.5 | 43.2 | |
| OVFormerVenue=BMVC’24, Prompt Tuning=false, Text Feature=DINOv22025.11 | 49.8 | 43.8 | 35.8 | 27.8 | 19.2 | 35.3 | |
| STOV-TALVenue=WACV’25, Prompt Tuning=false, Text Feature=CLIP-B2025.11 | 47.8 | 39.1 | 28.4 | 17.6 | 9.1 | 28.4 | |
| mProTEAVenue=TCSVT’24, Prompt Tuning=true, Text Feature=CLIP-B2025.11 | 43.1 | 38.2 | 28.2 | 18.1 | 8.7 | 27.9 | |
| STALETraining-based=true2024.04 | 40.5 | 32.3 | 23.5 | 15.3 | 7.6 | 23.8 | |
| STALETraining paradigm=training-based2026.05 | 40.5 | 32.3 | 23.5 | 15.3 | 7.6 | 23.8 | |
| STABLEVenue=ECCV’22, Prompt Tuning=true, Text Feature=CLIP-B2025.11 | 40.5 | 32.3 | 23.5 | 15.3 | 7.6 | 23.8 | |
| DeTALVenue=TPAMI’24, Prompt Tuning=true, Text Feature=CLIP-B2025.11 | 39.8 | 33.6 | 25.9 | 17.4 | 9.9 | 25.3 | |
| EffPromptTraining-based=true2024.04 | 39.7 | 31.6 | 23 | 14.9 | 7.5 | 23.3 | |
| EffPromptTraining paradigm=training-based2026.05 | 39.7 | 31.6 | 23 | 14.9 | 7.5 | 23.3 | |
| Eff-PromptVenue=ECCV’22, Prompt Tuning=true, Text Feature=CLIP-B2025.11 | 39.7 | 31.6 | 23 | 14.9 | 7.5 | 23.3 | |
| CLIP16 w/ DetectorBackbone=ViT-B/16, Training-based=true, Detector=Included2024.04 | 33 | 25.5 | 18.3 | 11.6 | 5.7 | 18.8 | |
| Detector + CLIPTraining paradigm=training-based2026.05 | 33 | 25.5 | 18.3 | 11.6 | 5.7 | 18.8 | |
| B-IVenue=-, Prompt Tuning=true, Text Feature=CLIP-B2025.11 | 33 | 25.5 | 18.3 | 11.6 | 5.7 | 18.8 | |
| B-IIVenue=-, Prompt Tuning=true, Text Feature=CLIP-B2025.11 | 28.5 | 20.3 | 17.1 | 10.5 | 6.9 | 16.6 | |
| TEGUTraining paradigm=non-training-based2026.05 | 23.3 | 15.7 | 9.1 | 4.7 | 2.4 | 11 | |
| VFEALSetting=online, Visual Encoder=ViCLIP, Optimization-Free=Yes, Out-of-Distribution=Yes2026.05 | 20.11 | 12.75 | 7.61 | 4.09 | 1.96 | 9.3 | |
| T3ALBackbone=ViT-L/14, Training-based=false2024.04 | 19.2 | 12.7 | 7.4 | 4.4 | 2.2 | 9.2 | |
| T3ALSetting=offline, Visual Encoder=CoCa, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 19.2 | 12.7 | 7.4 | 4.4 | 2.2 | 9.2 | |
| T3ALTraining paradigm=non-training-based2026.05 | 19.2 | 12.7 | 7.4 | 4.4 | 2.2 | 9.2 | |
| T3AL T=0Backbone=ViT-L/14, Training-based=false, Adaptation Steps=02024.04 | 11.1 | 6.5 | 3.2 | 1.5 | 0.6 | 4.6 | |
| T3ALT=0Setting=offline, Visual Encoder=CoCa, Optimization-Free=Yes, Out-of-Distribution=Yes2026.05 | 11.1 | 6.5 | 3.2 | 1.5 | 0.6 | 4.6 | |
| CoCaBackbone=ViT-L/14, Training-based=false2024.04 | 7.8 | 4.6 | 2.5 | 1.3 | 0.6 | 3.4 | |
| EffPromptSetting=offline, Visual Encoder=CLIP, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 7.1 | 5.9 | 4.5 | 3.4 | 2.2 | 4.6 | |
| CLIP16Backbone=ViT-B/16, Training-based=false2024.04 | 6.9 | 3.8 | 2.1 | 1.1 | 0.6 | 2.9 | |
| Baseline-IISetting=online, Visual Encoder=ViCLIP, Optimization-Free=Yes, Out-of-Distribution=Yes2026.05 | 6.54 | 4.16 | 2.96 | 2.26 | 1.54 | 3.49 | |
| Baseline-ISetting=online, Visual Encoder=CLIP, Optimization-Free=Yes, Out-of-Distribution=Yes2026.05 | 5.88 | 3.3 | 1.48 | 1.06 | 0.56 | 2.46 | |
| CLIP32Backbone=ViT-B/32, Training-based=false2024.04 | 5.5 | 3.3 | 1.9 | 0.9 | 0.4 | 2.4 | |
| MATR*Setting=online, Visual Encoder=TSN, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.86 | 0.79 | 0.76 | 0.52 | 0.33 | 0.65 | |
| HAT*Setting=online, Visual Encoder=TSN, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.78 | 0.75 | 0.62 | 0.53 | 0.31 | 0.6 | |
| STALESetting=offline, Visual Encoder=CLIP, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.5 | 0.3 | 0.2 | 0.2 | 0.2 | 0.3 | |
| MATR*Setting=online, Visual Encoder=ViCLIP, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.38 | 0.34 | 0.28 | 0.15 | 0.1 | 0.25 | |
| HAT*Setting=online, Visual Encoder=ViCLIP, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.33 | 0.28 | 0.18 | 0.09 | 0.04 | 0.18 |