SOTAVerified

Action Recognition

Action Recognition is a computer vision task that involves recognizing human actions in videos or images. The goal is to classify and categorize the actions being performed in the video or image into a predefined set of action classes.

In the video domain, it is an open question whether training an action classification network on a sufficiently large dataset, will give a similar boost in performance when applied to a different temporal task or dataset. The challenges of building video datasets has meant that most popular benchmarks for action recognition are small, having on the order of 10k videos.

Please note some benchmarks may be located in the Action Classification or Video Classification tasks, e.g. Kinetics-400.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1251–1300 of 2759 papers

TitleStatusHype
Transductive Universal Transport for Zero-Shot Action Recognition—0
Fundamental Limits of Transfer Learning in Binary Classifications—0
UniFormer: Unified Transformer for Efficient Spatial-Temporal Representation LearningCode1
Self-Supervised Learning of Motion-Informed Latents—0
Cross-Stage Transformer for Video Learning—0
Vi-MIX FOR SELF-SUPERVISED VIDEO REPRESENTATION—0
Information Elevation Network for Fast Online Action Detection—0
Fusion-GCN: Multimodal Action Recognition using Graph Convolutional NetworksCode1
Self-Supervised Video Representation Learning by Video Incoherence Detection—0
Long Short View Feature Decomposition via Contrastive Video Representation Learning—0
Unsupervised View-Invariant Human Posture Representation—0
ActionCLIP: A New Paradigm for Video Action RecognitionCode1
Adversarial Bone Length Attack on Action Recognition—0
Egocentric View Hand Action Recognition by Leveraging Hand Surface and Hand Grasp Type—0
Hierarchical Graph Convolutional Skeleton Transformer for Action Recognition—0
Improving Phenotype Prediction using Long-Range Spatio-Temporal Dynamics of Functional ConnectivityCode1
Efficient Action Recognition Using Confidence Distillation—0
Video Pose Distillation for Few-Shot, Fine-Grained Sports Action RecognitionCode1
Conditional Extreme Value Theory for Open Set Video Domain AdaptationCode0
LIGAR: Lightweight General-purpose Action Recognition—0
Learning Cross-modal Contrastive Features for Video Domain Adaptation—0
Shifted Chunk Transformer for Spatio-Temporal Representational Learning—0
BiaSwap: Removing dataset bias with bias-tailored swapping augmentation—0
Few Shot Activity Recognition Using Variational Inference—0
MM-ViT: Multi-Modal Video Transformer for Compressed Video Action Recognition—0
Blindly Assess Quality of In-the-Wild Videos via Quality-aware Pre-training and Motion PerceptionCode1
Self-Supervised Video Representation Learning with Meta-Contrastive Network—0
The Multi-Modal Video Reasoning and Analyzing Competition—0
Channel-Temporal Attention for First-Person Video Domain Adaptation—0
Learning Skeletal Graph Neural Networks for Hard 3D Pose EstimationCode0
Temporal Action Segmentation with High-level Complex Activity Labels—0
Few-Shot Fine-Grained Action Recognition via Bidirectional Attention and Contrastive Meta-LearningCode0
Spatio-Temporal Human Action Recognition Modelwith Flexible-interval Sampling and Normalization—0
Learning Visual Affordance Grounding from Demonstration Videos—0
Learning Multi-Granular Spatio-Temporal Graph Network for Skeleton-based Action RecognitionCode1
AutoVideo: An Automated Video Action Recognition SystemCode1
Skeleton-Contrastive 3D Action Representation LearningCode1
One-Shot Object Affordance Detection in the WildCode1
Temporal Action Localization Using Gated Recurrent UnitsCode0
Feature-Supervised Action Modality Transfer—0
Elaborative Rehearsal for Zero-shot Action RecognitionCode1
Unifying Nonlocal Blocks for Neural NetworksCode1
Skeleton Cloud Colorization for Unsupervised 3D Action Representation Learning—0
Classifying action correctness in physical rehabilitation exercises—0
Video Based Fall Detection Using Human PosesCode1
Lighter Stacked Hourglass Human Pose Estimation—0
A New Split for Evaluating True Zero-Shot Action RecognitionCode0
Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action RecognitionCode1
Adaptive Recursive Circle Framework for Fine-grained Action Recognition—0
TinyAction Challenge: Recognizing Real-world Low-resolution Activities in VideosCode1
Show:102550
← PrevPage 26 of 56Next →
All datasetsSomething-Something V2UCF101HMDB-51Something-Something V1AVA v2.2EPIC-KITCHENS-100NTU RGB+DNTU RGB+D 120Diving-48ActivityNetAVA v2.1H2O (2 Hands and Objects)

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1MViTv2-B (IN-21K + Kinetics400 pretrain)Top-5 Accuracy93.4—Unverified
2RSANet-R50 (8+16 frames, ImageNet pretrained, 2 clips)Top-5 Accuracy91.1—Unverified
3MVD (Kinetics400 pretrain, ViT-H, 16 frame)Top-1 Accuracy77.3—Unverified
4InternVideoTop-1 Accuracy77.2—Unverified
5DejaVidTop-1 Accuracy77.2—Unverified
6InternVideo2-1BTop-1 Accuracy77.1—Unverified
7VideoMAE V2-gTop-1 Accuracy77—Unverified
8MVD (Kinetics400 pretrain, ViT-L, 16 frame)Top-1 Accuracy76.7—Unverified
9Hiera-L (no extra data)Top-1 Accuracy76.5—Unverified
10TubeViT-LTop-1 Accuracy76.1—Unverified
#ModelMetricClaimedVerifiedStatus
1FTP-UniFormerV2-L/143-fold Accuracy99.7—Unverified
2OmniVec3-fold Accuracy99.6—Unverified
3VideoMAE V2-g3-fold Accuracy99.6—Unverified
4OmniVec23-fold Accuracy99.6—Unverified
5BIKE3-fold Accuracy98.8—Unverified
6SMART3-fold Accuracy98.64—Unverified
7ZeroI2V ViT-L/143-fold Accuracy98.6—Unverified
8PERF-Net (multi-distilled S3D)3-fold Accuracy98.6—Unverified
9OmniSource (SlowOnly-8x8-R101-RGB + I3D-Flow)3-fold Accuracy98.6—Unverified
10Text4Vis3-fold Accuracy98.2—Unverified