Temporal Action Localization on THUMOS14 v1.0 (50%-50%)
48.8mAP (Avg)STOV-TAL
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| STOV-TALTraining=supervised, VAE=false2026.01 | 48.8 | — | — | — | — | — | |
| PyraTokTraining=zero-shot, VAE=true2026.01 | 33.17 | — | — | — | — | — | |
| STOV-TALTraining=zero-shot, VAE=false2026.01 | 31.5 | — | — | — | — | — | |
| ZEETADTraining paradigm=training-based2026.05 | 30.2 | 45.2 | 38.8 | 30.8 | 22.5 | 13.7 | |
| LARPTraining=zero-shot, VAE=true2026.01 | 27.42 | — | — | — | — | — | |
| SweetTokTraining=zero-shot, VAE=true2026.01 | 25.32 | — | — | — | — | — | |
| DeTALTraining=supervised, VAE=false2026.01 | 24.1 | — | — | — | — | — | |
| OmniTokenizerTraining=zero-shot, VAE=true2026.01 | 23.47 | — | — | — | — | — | |
| VideoVae+Training=zero-shot, VAE=true2026.01 | 23.12 | — | — | — | — | — | |
| STALETraining=supervised, VAE=false2026.01 | 22.2 | — | — | — | — | — | |
| STALEApproach=training-based2024.04 | 22.2 | 38.3 | 30.7 | 21.2 | 13.8 | 7 | |
| STALETraining paradigm=training-based2026.05 | 22.2 | 38.3 | 30.7 | 21.2 | 13.8 | 7 | |
| EffPromptApproach=training-based2024.04 | 21.9 | 37.2 | 29.6 | 21.6 | 14 | 7.2 | |
| EffPromptTraining paradigm=training-based2026.05 | 21.9 | 37.2 | 29.6 | 21.6 | 14 | 7.2 | |
| CLIP16 w/ DetectorBackbone=ViT-B/16, Approach=training-based2024.04 | 15.7 | 27.2 | 21.3 | 15.3 | 9.7 | 4.8 | |
| Detector + CLIPTraining paradigm=training-based2026.05 | 15.7 | 27.2 | 21.3 | 15.3 | 9.7 | 4.8 | |
| TEGUTraining paradigm=non-training-based2026.05 | 10.5 | 22.4 | 14.8 | 8.6 | 4.3 | 2.2 | |
| T3ALT=502024.04 | 10.4 | 20.7 | 14.3 | 8.9 | 5.3 | 2.7 | |
| T3ALSetting=offline, Visual Encoder=CoCa, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 10.4 | 20.7 | 14.3 | 8.9 | 5.3 | 2.7 | |
| T3ALTraining paradigm=non-training-based2026.05 | 10.4 | 20.7 | 14.3 | 8.9 | 5.3 | 2.7 | |
| VFEALSetting=online, Visual Encoder=ViCLIP, Optimization-Free=Yes, Out-of-Distribution=Yes2026.05 | 9.24 | 19.51 | 12.62 | 7.71 | 4.31 | 2.05 | |
| T3ALT=02024.04 | 4.8 | 11.4 | 6.8 | 3.5 | 1.7 | 0.6 | |
| T3ALT=0Setting=offline, Visual Encoder=CoCa, Optimization-Free=Yes, Out-of-Distribution=Yes2026.05 | 4.8 | 11.4 | 6.8 | 3.5 | 1.7 | 0.6 | |
| EffPromptSetting=offline, Visual Encoder=CLIP, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 3.6 | 5.4 | 4.4 | 3.5 | 2.7 | 1.9 | |
| CoCaBackbone=ViT-L/142024.04 | 3.5 | 8.4 | 4.7 | 2.5 | 1.2 | 0.6 | |
| Baseline-IISetting=online, Visual Encoder=ViCLIP, Optimization-Free=Yes, Out-of-Distribution=Yes2026.05 | 3.23 | 6.19 | 4.01 | 2.64 | 2.06 | 1.23 | |
| CLIP16Backbone=ViT-B/162024.04 | 3.1 | 7.5 | 4.2 | 2.2 | 1.1 | 0.6 | |
| CLIP32Backbone=ViT-B/322024.04 | 3 | 7.2 | 4.1 | 2.3 | 1.1 | 0.5 | |
| Baseline-ISetting=online, Visual Encoder=CLIP, Optimization-Free=Yes, Out-of-Distribution=Yes2026.05 | 2.6 | 6.21 | 3.44 | 1.81 | 1.03 | 0.52 | |
| STALESetting=offline, Visual Encoder=CLIP, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.7 | 1.3 | 0.7 | 0.6 | 0.6 | 0.4 | |
| MATR*Setting=online, Visual Encoder=TSN, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.54 | 0.75 | 0.66 | 0.56 | 0.45 | 0.26 | |
| HAT*Setting=online, Visual Encoder=TSN, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.46 | 0.56 | 0.53 | 0.49 | 0.41 | 0.32 | |
| MATR*Setting=online, Visual Encoder=ViCLIP, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.21 | 0.38 | 0.31 | 0.19 | 0.14 | 0.05 | |
| HAT*Setting=online, Visual Encoder=ViCLIP, Optimization-Free=No, Out-of-Distribution=Yes2026.05 | 0.16 | 0.27 | 0.22 | 0.15 | 0.11 | 0.05 |