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 301–350 of 2759 papers

TitleStatusHype
Complex Sequential Understanding through the Awareness of Spatial and Temporal ConceptsCode1
EAN: Event Adaptive Network for Enhanced Action RecognitionCode1
BABEL: Bodies, Action and Behavior with English LabelsCode1
EgoExOR: An Ego-Exo-Centric Operating Room Dataset for Surgical Activity UnderstandingCode1
EgoSurgery-Phase: A Dataset of Surgical Phase Recognition from Egocentric Open Surgery VideosCode1
Collaborating Domain-shared and Target-specific Feature Clustering for Cross-domain 3D Action RecognitionCode1
Encoding Surgical Videos as Latent Spatiotemporal Graphs for Object and Anatomy-Driven ReasoningCode1
End-to-End Learning of Visual Representations from Uncurated Instructional VideosCode1
Actor-Context-Actor Relation Network for Spatio-Temporal Action LocalizationCode1
Enhancing Unsupervised Video Representation Learning by Decoupling the Scene and the MotionCode1
Action-Conditioned 3D Human Motion Synthesis with Transformer VAECode1
Compressing Recurrent Neural Networks with Tensor Ring for Action RecognitionCode1
Epic-Sounds: A Large-scale Dataset of Actions That SoundCode1
Mitigating and Evaluating Static Bias of Action Representations in the Background and the ForegroundCode1
EventRPG: Event Data Augmentation with Relevance Propagation GuidanceCode1
Event Stream based Human Action Recognition: A High-Definition Benchmark Dataset and AlgorithmsCode1
ExACT: Language-guided Conceptual Reasoning and Uncertainty Estimation for Event-based Action Recognition and MoreCode1
CoCon: Cooperative-Contrastive LearningCode1
CMD: Self-supervised 3D Action Representation Learning with Cross-modal Mutual DistillationCode1
CoFInAl: Enhancing Action Quality Assessment with Coarse-to-Fine Instruction AlignmentCode1
Computer Vision for Clinical Gait Analysis: A Gait Abnormality Video DatasetCode1
Multi-Granularity Hand Action DetectionCode1
CIAGAN: Conditional Identity Anonymization Generative Adversarial NetworksCode1
Florence: A New Foundation Model for Computer VisionCode1
CHASE: Learning Convex Hull Adaptive Shift for Skeleton-based Multi-Entity Action RecognitionCode1
Frequency Guidance Matters: Skeletal Action Recognition by Frequency-Aware Mixed TransformerCode1
CIDEr: Consensus-based Image Description EvaluationCode1
GCN-DevLSTM: Path Development for Skeleton-Based Action RecognitionCode1
Attention Prompt Tuning: Parameter-efficient Adaptation of Pre-trained Models for Spatiotemporal ModelingCode1
3DV: 3D Dynamic Voxel for Action Recognition in Depth VideoCode1
Anonymization for Skeleton Action RecognitionCode1
Grad-CAM++: Improved Visual Explanations for Deep Convolutional NetworksCode1
Graph in Graph Neural NetworkCode1
Graph Convolution with Low-rank Learnable Local FiltersCode1
3DYoga90: A Hierarchical Video Dataset for Yoga Pose UnderstandingCode1
Benchmarking Micro-action Recognition: Dataset, Methods, and ApplicationsCode1
Hear Me Out: Fusional Approaches for Audio Augmented Temporal Action LocalizationCode1
Hidden Two-Stream Convolutional Networks for Action RecognitionCode1
Audio-Visual Instance Discrimination with Cross-Modal AgreementCode1
Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action RecognitionCode1
A Unified Multimodal De- and Re-coupling Framework for RGB-D Motion RecognitionCode1
Augmented Neural Fine-Tuning for Efficient Backdoor PurificationCode1
Augmented Skeleton Based Contrastive Action Learning with Momentum LSTM for Unsupervised Action RecognitionCode1
Holistic Interaction Transformer Network for Action DetectionCode1
Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action RecognitionCode1
Home Action Genome: Cooperative Compositional Action UnderstandingCode1
CLIP-guided Prototype Modulating for Few-shot Action RecognitionCode1
Hybrid Relation Guided Set Matching for Few-shot Action RecognitionCode1
A Dense-Sparse Complementary Network for Human Action Recognition based on RGB and Skeleton ModalitiesCode1
Concatenated Masked Autoencoders as Spatial-Temporal LearnerCode1
Show:102550
← PrevPage 7 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