SOTAVerified

Activity Recognition

Human Activity Recognition is the problem of identifying events performed by humans given a video input. It is formulated as a binary (or multiclass) classification problem of outputting activity class labels. Activity Recognition is an important problem with many societal applications including smart surveillance, video search/retrieval, intelligent robots, and other monitoring systems.

Source: Learning Latent Sub-events in Activity Videos Using Temporal Attention Filters

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 726–750 of 1322 papers

TitleStatusHype
Human-like Relational Models for Activity Recognition in Video—0
A Light-weight Deep Human Activity Recognition Algorithm Using Multi-knowledge Distillation—0
Early Mobility Recognition for Intensive Care Unit Patients Using Accelerometers—0
Reducing numerical precision preserves classification accuracy in Mondrian ForestsCode0
Human Activity Recognition using Continuous Wavelet Transform and Convolutional Neural NetworksCode0
PALMAR: Towards Adaptive Multi-inhabitant Activity Recognition in Point-Cloud Technology—0
A Survey on Human-aware Robot Navigation—0
A compressive multi-kernel method for privacy-preserving machine learning—0
Multi-Modal Prototype Learning for Interpretable Multivariable Time Series Classification—0
Privacy-Preserving Eye-tracking Using Deep Learning—0
Long Term Object Detection and Tracking in Collaborative Learning Environments—0
FedHealth 2: Weighted Federated Transfer Learning via Batch Normalization for Personalized Healthcare—0
Similarity Embedding Networks for Robust Human Activity Recognition—0
Meta-HAR: Federated Representation Learning for Human Activity RecognitionCode1
Quantization and Deployment of Deep Neural Networks on MicrocontrollersCode0
Explainable Activity Recognition for Smart Home Systems—0
Egocentric Activity Recognition and Localization on a 3D Map—0
Social Behaviour Understanding using Deep Neural Networks: Development of Social Intelligence Systems—0
ASM2TV: An Adaptive Semi-Supervised Multi-Task Multi-View Learning Framework for Human Activity RecognitionCode0
Learning Group Activities from Skeletons without Individual Action LabelsCode1
Event-LSTM: An Unsupervised and Asynchronous Learning-based Representation for Event-based Data—0
Evaluating Deep Neural Network Ensembles by Majority Voting cum Meta-Learning scheme—0
Human Activity Recognition Models in Ontology Networks—0
Activity-Aware Deep Cognitive Fatigue Assessment using Wearables—0
Three-stream network for enriched Action Recognition—0
Show:102550
← PrevPage 30 of 53Next →
All datasetsRWF-2000Stanford40First-Person Hand Action BenchmarkSelf-stimulatory Behavior Dataset

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Structured Keypoint PoolingAccuracy93.4—Unverified
2Semi-Supervised Hard Attention (SSHA); pretrained on Deepmind Kinetics datasetAccuracy90.4—Unverified
3Human Skeletons + Change DetectionAccuracy90.25—Unverified
4Separable Convolutional LSTMAccuracy89.75—Unverified
5SPIL ConvolutionAccuracy89.3—Unverified
6Flow Gated NetworkAccuracy87.25—Unverified
#ModelMetricClaimedVerifiedStatus
1FocusCLIPTop-3 Accuracy (%)10.47—Unverified
2CLIPTop-3 Accuracy (%)6.49—Unverified
#ModelMetricClaimedVerifiedStatus
1Boutaleb et al.1:1 Accuracy97.91—Unverified
#ModelMetricClaimedVerifiedStatus
1all-landmark-modelActivity Recognition0.76—Unverified