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 2401–2450 of 2759 papers

TitleStatusHype
Skeleton-based Action Recognition Using LSTM and CNN—0
Learning Human Pose Models from Synthesized Data for Robust RGB-D Action Recognition—0
Zero-Shot Action Recognition With Error-Correcting Output Codes—0
Modeling Sub-Event Dynamics in First-Person Action Recognition—0
Deep Sequential Context Networks for Action Prediction—0
Multi-Task Clustering of Human Actions by Sharing Information—0
Spatio-Temporal Vector of Locally Max Pooled Features for Action Recognition in Videos—0
Global Context-Aware Attention LSTM Networks for 3D Action Recognition—0
Spatio-Temporal Naive-Bayes Nearest-Neighbor (ST-NBNN) for Skeleton-Based Action RecognitionCode0
Temporal Residual Networks for Dynamic Scene RecognitionCode0
Joint Discriminative Bayesian Dictionary and Classifier Learning—0
Binary Coding for Partial Action Analysis With Limited Observation Ratios—0
Alternative Semantic Representations for Zero-Shot Human Action Recognition—0
Recurrent Residual Learning for Action Recognition—0
Skeleton-Based Action Recognition Using Spatio-Temporal LSTM Network with Trust Gates—0
Detekcja upadku i wybranych akcji na sekwencjach obrazów cyfrowych—0
Learning without Prejudice: Avoiding Bias in Webly-Supervised Action Recognition—0
Joint Max Margin and Semantic Features for Continuous Event Detection in Complex Scenes—0
Learning to Learn from Noisy Web Videos—0
Concurrence-Aware Long Short-Term Sub-Memories for Person-Person Action Recognition—0
Deep manifold-to-manifold transforming network for action recognition—0
Continuous Video to Simple Signals for Swimming Stroke Detection with Convolutional Neural Networks—0
Sequence Summarization Using Order-constrained Kernelized Feature Subspaces—0
Ridiculously Fast Shot Boundary Detection with Fully Convolutional Neural NetworksCode0
Two-Stream 3D Convolutional Neural Network for Skeleton-Based Action Recognition—0
Learning Spatiotemporal Features for Infrared Action Recognition with 3D Convolutional Neural Networks—0
Single Image Action Recognition by Predicting Space-Time Saliency—0
An Improved Video Analysis using Context based Extension of LSH—0
CHAM: action recognition using convolutional hierarchical attention model—0
Deep Spatio-temporal Manifold Network for Action Recognition—0
Investigation of Different Skeleton Features for CNN-based 3D Action RecognitionCode0
Joint Denoising / Compression of Image Contours via Shape Prior and Context Tree—0
Action Understanding with Multiple Classes of Actors—0
Body Joint guided 3D Deep Convolutional Descriptors for Action Recognition—0
An Analysis of Action Recognition Datasets for Language and Vision Tasks—0
Second-order Temporal Pooling for Action Recognition—0
Context-based Object Viewpoint Estimation: A 2D Relational Approach—0
Invariant recognition drives neural representations of action sequences—0
Skeleton based action recognition using translation-scale invariant image mapping and multi-scale deep cnn—0
Skeleton Boxes: Solving skeleton based action detection with a single deep convolutional neural network—0
Interpretable 3D Human Action Analysis with Temporal Convolutional NetworksCode0
Learning to Estimate Pose by Watching VideosCode0
UC Merced Submission to the ActivityNet Challenge 2016—0
Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks—0
First-Person Hand Action Benchmark with RGB-D Videos and 3D Hand Pose AnnotationsCode0
Generalized Rank Pooling for Activity Recognition—0
Action Representation Using Classifier Decision Boundaries—0
Two Stream LSTM: A Deep Fusion Framework for Human Action Recognition—0
Chained Multi-stream Networks Exploiting Pose, Motion, and Appearance for Action Classification and DetectionCode0
Unsupervised Action Proposal Ranking through Proposal Recombination—0
Show:102550
← PrevPage 49 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