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

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 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
Talk is Not Always Cheap: Promoting Wireless Sensing Models with Text PromptsCode0
Chain-of-Thought Textual Reasoning for Few-shot Temporal Action Localization—0
PCBEAR: Pose Concept Bottleneck for Explainable Action Recognition—0
OwlSight: A Robust Illumination Adaptation Framework for Dark Video Human Action Recognition—0
Temporal-Guided Spiking Neural Networks for Event-Based Human Action Recognition—0
Towards Achieving Perfect Multimodal Alignment—0
TimeLoc: A Unified End-to-End Framework for Precise Timestamp Localization in Long VideosCode1
Semi-Supervised Audio-Visual Video Action Recognition with Audio Source Localization Guided Mixup—0
A Simple and Efficient Baseline for Video Action Recognition—0
SNN-Driven Multimodal Human Action Recognition via Event Camera and Skeleton Data Fusion—0
HyLiFormer: Hyperbolic Linear Attention for Skeleton-based Human Action Recognition—0
Minimalistic Video Saliency Prediction via Efficient Decoder & Spatio Temporal Action Cues—0
XRF V2: A Dataset for Action Summarization with Wi-Fi Signals, and IMUs in Phones, Watches, Earbuds, and GlassesCode1
SMART-Vision: Survey of Modern Action Recognition Techniques in Vision—0
Rethinking Pseudo-Label Guided Learning for Weakly Supervised Temporal Action Localization from the Perspective of Noise Correction—0
Visual WetlandBirds Dataset: Bird Species Identification and Behavior Recognition in VideosCode0
Heterogeneous Skeleton-Based Action Representation Learning—0
Boosting Point-Supervised Temporal Action Localization through Integrating Query Reformation and Optimal Transport—0
Semantic-guided Cross-Modal Prompt Learning for Skeleton-based Zero-shot Action Recognition—0
Action Detail Matters: Refining Video Recognition with Local Action Queries—0
LoKi: Low-dimensional KAN for Efficient Fine-tuning Image Models—0
Weakly Supervised Temporal Action Localization via Dual-Prior Collaborative Learning Guided by Multimodal Large Language Models—0
MODA: Motion-Drift Augmentation for Inertial Human Motion Analysis—0
DAVE: Diverse Atomic Visual Elements Dataset with High Representation of Vulnerable Road Users in Complex and Unpredictable Environments—0
Generalized Uncertainty-Based Evidential Fusion with Hybrid Multi-Head Attention for Weak-Supervised Temporal Action LocalizationCode0
Video Domain Incremental Learning for Human Action Recognition in Home Environments—0
Future Aspects in Human Action Recognition: Exploring Emerging Techniques and Ethical Influences—0
Temporal Action Localization with Cross Layer Task Decoupling and RefinementCode1
SkelMamba: A State Space Model for Efficient Skeleton Action Recognition of Neurological Disorders—0
An End-to-End Two-Stream Network Based on RGB Flow and Representation Flow for Human Action Recognition—0
OccludeNet: A Causal Journey into Mixed-View Actor-Centric Video Action Recognition under OcclusionsCode0
Multilevel semantic and adaptive actionness learning for weakly supervised temporal action localizationCode0
Adaptive Hyper-Graph Convolution Network for Skeleton-based Human Action Recognition with Virtual Connections—0
When Spatial meets Temporal in Action Recognition—0
Efficient Transfer Learning for Video-language Foundation ModelsCode0
Can MLLMs Guide Weakly-Supervised Temporal Action Localization Tasks?—0
Extended multi-stream temporal-attention module for skeleton-based human action recognition (HAR)—0
Human Action Recognition (HAR) Using Skeleton-based Spatial Temporal Relative Transformer Network: ST-RTR—0
Unsupervised Domain Adaptation for Action Recognition via Self-Ensembling and Conditional Embedding Alignment—0
SpikMamba: When SNN meets Mamba in Event-based Human Action RecognitionCode1
Show:102550
← PrevPage 1 of 30Next →
All datasetsTHUMOS14ActivityNet-1.3HACSFineActionMultiTHUMOSCrossTaskEPIC-KITCHENS-100MUSESActivityNet-1.2Ego4D MQ testEgo4D MQ valMEXaction2

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AdaTAD (VideoMAEv2-giant)Avg mAP (0.3:0.7)76.9—Unverified
2RDFA-S6 (InternVideo2-6B)Avg mAP (0.3:0.7)74.2—Unverified
3ActionMamba(InternVideo2-6B)Avg mAP (0.3:0.7)72.72—Unverified
4GCMmAP IOU@0.172.5—Unverified
5AGT (Ours)mAP IOU@0.172.1—Unverified
6InternVideo2-6BAvg mAP (0.3:0.7)72—Unverified
7ActionFormer (InternVideo features)Avg mAP (0.3:0.7)71.58—Unverified
8TriDet (VideoMAE v2-g feature)Avg mAP (0.3:0.7)70.1—Unverified
9InternVideo2-1BAvg mAP (0.3:0.7)69.8—Unverified
10ActionFormer (VideoMAE V2-g features)Avg mAP (0.3:0.7)69.6—Unverified
#ModelMetricClaimedVerifiedStatus
1UnLoc-LmAP IOU@0.559.3—Unverified
2RDFA-S6 (InternVideo2-6B)mAP42.9—Unverified
3ActionMamba (InternVideo2-6B)mAP42.02—Unverified
4PRN+BMN (ensemble)mAP42—Unverified
5AdaTAD (VideoMAEv2-giant)mAP41.93—Unverified
6InternVideo2-6BmAP41.2—Unverified
7InternVideo2-1BmAP40.4—Unverified
8UniMD+Sync.mAP39.83—Unverified
9PRN (CSN)mAP39.4—Unverified
10InternVideomAP39—Unverified
#ModelMetricClaimedVerifiedStatus
1RDFA-S6 (InternVideo2-6B)Average-mAP45.8—Unverified
2ActionMamba(InternVideo2-6B)Average-mAP44.56—Unverified
3DyFADet(VideoMAEv2)Average-mAP44.3—Unverified
4InternVideo2-6BAverage-mAP43.3—Unverified
5TriDet (VideoMAEv2)Average-mAP43.1—Unverified
6InternVideo2-1BAverage-mAP42.4—Unverified
7InternVideoAverage-mAP41.55—Unverified
8TriDet (SlowFast)Average-mAP38.6—Unverified
9TriDet (I3D RGB)Average-mAP36.8—Unverified
10TadTr (I3D RGB)Average-mAP32.09—Unverified
#ModelMetricClaimedVerifiedStatus
1RDFA-S6 (InternVideo2-6B)mAP29.6—Unverified
2ActionMamba(InternVideo2-6B)mAP29.04—Unverified
3InternVideo2-6BmAP27.7—Unverified
4DyFADet (VideoMAE v2-g)mAP23.8—Unverified
5VideoMAE V2-gmAP18.24—Unverified
6InternVideomAP17.57—Unverified
7BMN (i3d feaure)mAP9.25—Unverified
8G-TAD (i3d feature)mAP9.06—Unverified
9DBG (i3d feature)mAP6.75—Unverified
#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
#ModelMetricClaimedVerifiedStatus
1VideoCLIPRecall47.3—Unverified
2VLMRecall46.5—Unverified
3TACoRecall42.5—Unverified
4Text-Video EmbeddingRecall33.6—Unverified
5Fully-supervised upper-boundRecall31.6—Unverified
6ZhukovRecall22.4—Unverified
7AlayracRecall13.3—Unverified
#ModelMetricClaimedVerifiedStatus
1AdaTAD (verb, VideoMAE-L)Avg mAP (0.1-0.5)29.3—Unverified
2TriDet (verb)Avg mAP (0.1-0.5)25.4—Unverified
3TemporalMaxer (verb)Avg mAP (0.1-0.5)24.5—Unverified
4ActionFormer (verb)Avg mAP (0.1-0.5)23.5—Unverified
5G-TAD (verb)Avg mAP (0.1-0.5)9.4—Unverified
6BMN (verb)Avg mAP (0.1-0.5)8.4—Unverified
#ModelMetricClaimedVerifiedStatus
1TemporalMaxermAP27.2—Unverified
2MUSESmAP18.6—Unverified
#ModelMetricClaimedVerifiedStatus
1DeepMetricLearnermAP IOU@0.535.2—Unverified
#ModelMetricClaimedVerifiedStatus
1ActionFormer (SlowFast+Omnivore+EgoVLP)Average mAP21.76—Unverified
#ModelMetricClaimedVerifiedStatus
1ActionFormer (SlowFast+Omnivore+EgoVLP)Average mAP21.4—Unverified
#ModelMetricClaimedVerifiedStatus
1S-CNNmAP7.4—Unverified