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 101–125 of 1322 papers

TitleStatusHype
Self-supervised transfer learning of physiological representations from free-living wearable dataCode1
Learning Generalizable Physiological Representations from Large-scale Wearable DataCode1
HHAR-net: Hierarchical Human Activity Recognition using Neural NetworksCode1
Improved Actor Relation Graph based Group Activity RecognitionCode1
Skeleton-based Action Recognition via Spatial and Temporal Transformer NetworksCode1
DANA: Dimension-Adaptive Neural Architecture for Multivariate Sensor DataCode1
SeCo: Exploring Sequence Supervision for Unsupervised Representation LearningCode1
3D Human Shape and Pose from a Single Low-Resolution Image with Self-Supervised LearningCode1
ESPRESSO: Entropy and ShaPe awaRe timE-Series SegmentatiOn for processing heterogeneous sensor dataCode1
Attention-Based Deep Learning Framework for Human Activity Recognition with User AdaptationCode1
SleepPoseNet: Multi-View Learning for Sleep Postural Transition Recognition Using UWBCode1
Human Activity Recognition from Wearable Sensor Data Using Self-AttentionCode1
Gimme Signals: Discriminative signal encoding for multimodal activity recognitionCode1
Convolutional Tensor-Train LSTM for Spatio-temporal LearningCode1
Privacy and Utility Preserving Sensor-Data TransformationsCode1
NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity UnderstandingCode1
Mobile Sensor Data AnonymizationCode1
Semi-Supervised Online Structure Learning for Composite Event RecognitionCode1
Protecting Sensory Data against Sensitive InferencesCode1
Multivariate LSTM-FCNs for Time Series ClassificationCode1
Real-world Anomaly Detection in Surveillance VideosCode1
DeepSense: A Unified Deep Learning Framework for Time-Series Mobile Sensing Data ProcessingCode1
OSL𝛼: Online Structure Learning Using Background Knowledge AxiomatizationCode1
ZKP-FedEval: Verifiable and Privacy-Preserving Federated Evaluation using Zero-Knowledge Proofsβ€”0
SEZ-HARN: Self-Explainable Zero-shot Human Activity Recognition NetworkCode0
Show:102550
← PrevPage 5 of 53Next β†’
All datasetsRWF-2000Stanford40First-Person Hand Action BenchmarkSelf-stimulatory Behavior Dataset

Benchmark Results

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