SOTAVerified

Temporal Action Localization

Temporal Action Localization aims to detect activities in the video stream and output beginning and end timestamps. It is closely related to Temporal Action Proposal Generation.

Papers

Showing 1–10 of 1477 papers

TitleStatusHype
DVFL-Net: A Lightweight Distilled Video Focal Modulation Network for Spatio-Temporal Action RecognitionCode0
Including Semantic Information via Word Embeddings for Skeleton-based Action Recognition—0
Zero-Shot Temporal Interaction Localization for Egocentric VideosCode1
A Review on Coarse to Fine-Grained Animal Action Recognition—0
CLIP-AE: CLIP-assisted Cross-view Audio-Visual Enhancement for Unsupervised Temporal Action Localization—0
DeepConvContext: A Multi-Scale Approach to Timeseries Classification in Human Activity RecognitionCode0
ProTAL: A Drag-and-Link Video Programming Framework for Temporal Action Localization—0
Are Spatial-Temporal Graph Convolution Networks for Human Action Recognition Over-Parameterized?Code0
Action Spotting and Precise Event Detection in Sports: Datasets, Methods, and Challenges—0
Bridge the Gap: From Weak to Full Supervision for Temporal Action Localization with PseudoFormer—0
Show:102550
← PrevPage 1 of 148Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1TriDet (VideoMAEv2)Average mAP37.5—Unverified
2DualDETR (I3D-rgb)Average mAP32.64—Unverified
3TriDet (I3D-rgb)Average mAP30.7—Unverified
4TemporalMaxerAverage mAP29.9—Unverified
5PointTADAverage mAP23.5—Unverified
6PDANAverage mAP17.3—Unverified
7MS-TCTAverage mAP16.2—Unverified
8MLADAverage mAP14.2—Unverified