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 1951–2000 of 2759 papers

TitleStatusHype
Human activity recognition from skeleton posesCode0
Action recognition with spatial-temporal discriminative filter banks—0
Cross-Enhancement Transform Two-Stream 3D ConvNets for Action Recognition—0
Gradient Weighted Superpixels for Interpretability in CNNs—0
Bypass Enhancement RGB Stream Model for Pedestrian Action Recognition of Autonomous Vehicles—0
I3D-LSTM: A New Model for Human Action RecognitionCode0
Sparse Coding of Shape Trajectories for Facial Expression and Action Recognition—0
STM: SpatioTemporal and Motion Encoding for Action Recognition—0
Discriminating Spatial and Temporal Relevance in Deep Taylor Decompositions for Explainable Activity RecognitionCode0
Action Recognition in Untrimmed Videos with Composite Self-Attention Two-Stream Framework—0
An Evaluation of Action Recognition Models on EPIC-KitchensCode1
Fitting, Comparison, and Alignment of Trajectories on Positive Semi-Definite Matrices with Application to Action Recognition—0
Two-Stream Video Classification with Cross-Modality Attention—0
Multi-Agent Reinforcement Learning Based Frame Sampling for Effective Untrimmed Video Recognition—0
Deep Multi-Kernel Convolutional LSTM Networks and an Attention-Based Mechanism for Videos—0
Open Set Domain Adaptation for Image and Action RecognitionCode0
SkeleMotion: A New Representation of Skeleton Joint Sequences Based on Motion Information for 3D Action RecognitionCode0
Unsupervised Learning for Optical Flow Estimation Using Pyramid Convolution LSTMCode0
A Novel Approach for Robust Multi Human Action Recognition and Summarization based on 3D Convolutional Neural Networks—0
Weakly Supervised Recognition of Surgical Gestures—0
Learning Visual Actions Using Multiple Verb-Only LabelsCode0
Motion-Aware Feature for Improved Video Anomaly Detection—0
Make Skeleton-based Action Recognition Model Smaller, Faster and BetterCode1
BMN: Boundary-Matching Network for Temporal Action Proposal GenerationCode1
Domain-Specific Priors and Meta Learning for Few-Shot First-Person Action Recognition—0
TARN: Temporal Attentive Relation Network for Few-Shot and Zero-Shot Action Recognition—0
Real-Time Driver State Monitoring Using a CNN Based Spatio-Temporal Approach—0
I Know the Relationships: Zero-Shot Action Recognition via Two-Stream Graph Convolutional Networks and Knowledge GraphsCode0
Spatiotemporal graph routing for skeleton-based action recognition—0
A Unified Deep Framework for Joint 3D Pose Estimation and Action Recognition from a Single RGB Camera—0
Slow Feature Analysis for Human Action Recognition—0
Two-stream Spatiotemporal Feature for Video QA Task—0
Video Action Recognition Via Neural Architecture Searching—0
Learning Shape-Motion Representations from Geometric Algebra Spatio-Temporal Model for Skeleton-Based Action Recognition—0
A Deep Learning Approach for Real-Time 3D Human Action Recognition from Skeletal Data—0
Non-Local Graph Convolutional Networks for Skeleton-Based Action RecognitionCode0
Deformable Tube Network for Action Detection in Videos—0
An Analysis of Deep Neural Networks with Attention for Action Recognition from a Neurophysiological Perspective—0
Few-Shot Video Classification via Temporal Alignment—0
A Comparative Review of Recent Kinect-based Action Recognition AlgorithmsCode0
Baidu-UTS Submission to the EPIC-Kitchens Action Recognition Challenge 2019—0
FBK-HUPBA Submission to the EPIC-Kitchens 2019 Action Recognition Challenge—0
R-STAN: Residual Spatial-Temporal Attention Network for Action Recognition—0
An Action Recognition network for specific target based on rMC and RPN—0
Unsupervised Learning of Object Structure and Dynamics from Videos—0
Spatio-Temporal Fusion Networks for Action Recognition—0
Towards Real-Time Action Recognition on Mobile Devices Using Deep Models—0
A Temporal Sequence Learning for Action Recognition and Prediction—0
Three-Stream Convolutional Neural Network With Multi-Task and Ensemble Learning for 3D Action Recognition—0
Accelerating temporal action proposal generation via high performance computing—0
Show:102550
← PrevPage 40 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
4DejaVidTop-1 Accuracy77.2—Unverified
5InternVideoTop-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
2OmniVec23-fold Accuracy99.6—Unverified
3VideoMAE V2-g3-fold Accuracy99.6—Unverified
4OmniVec3-fold Accuracy99.6—Unverified
5BIKE3-fold Accuracy98.8—Unverified
6SMART3-fold Accuracy98.64—Unverified
7OmniSource (SlowOnly-8x8-R101-RGB + I3D-Flow)3-fold Accuracy98.6—Unverified
8PERF-Net (multi-distilled S3D)3-fold Accuracy98.6—Unverified
9ZeroI2V ViT-L/143-fold Accuracy98.6—Unverified
10LGD-3D Two-stream3-fold Accuracy98.2—Unverified