Activity Recognition
254 papers with code • 4 benchmarks • 29 datasets
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
Benchmarks
These leaderboards are used to track progress in Activity Recognition
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
Libraries
Use these libraries to find Activity Recognition models and implementationsDatasets
GazeFollow
MotionSense
Drive&Act
RWF-2000
First-Person Hand Action Benchmark
MoVi
MSR ActionPairs
Home Action Genome
Subtasks
Latest papers
Quantized Distillation: Optimizing Driver Activity Recognition Models for Resource-Constrained Environments
calvintanama/qd-driver-activity-reco
•
•
The framework enhances 3D MobileNet, a neural architecture optimized for speed in video classification, by incorporating knowledge distillation and model quantization to balance model accuracy and computational efficiency.
Optimization-Free Test-Time Adaptation for Cross-Person Activity Recognition
However, the high computational cost of optimization-based TTA algorithms makes it intractable to run on resource-constrained edge devices.
Finding Order in Chaos: A Novel Data Augmentation Method for Time Series in Contrastive Learning
eth-siplab/Finding_Order_in_Chaos
•
•
NeurIPS 2023
The success of contrastive learning is well known to be dependent on data augmentation.
Human Activity Segmentation Challenge @ ECML/PKDD’23
Despite its importance, existing methods demonstrate limited efficacy on real-world multivariate time series data.
Hard No-Box Adversarial Attack on Skeleton-Based Human Action Recognition with Skeleton-Motion-Informed Gradient
luyg45/hardnoboxattack
•
•
ICCV 2023
Specifically, we first learn a motion manifold where we define an adversarial loss to compute a new gradient for the attack, named skeleton-motion-informed (SMI) gradient.
Evaluating Spiking Neural Network On Neuromorphic Platform For Human Activity Recognition
Energy efficiency and low latency are crucial requirements for designing wearable AI-empowered human activity recognition systems, due to the hard constraints of battery operations and closed-loop feedback.
MyDigitalFootprint: an extensive context dataset for pervasive computing applications at the edge
To demonstrate the dataset's effectiveness, we present three context-aware applications utilizing various machine learning tasks: (i) a social link prediction algorithm based on physical proximity data, (ii) daily-life activity recognition using smartphone-embedded sensors data, and (iii) a pervasive context-aware recommender system.
Combining Public Human Activity Recognition Datasets to Mitigate Labeled Data Scarcity
getalp/smartcomp2023-har-supervised-pretraining
•
•
In this paper, we propose a novel strategy to combine publicly available datasets with the goal of learning a generalized HAR model that can be fine-tuned using a limited amount of labeled data on an unseen target domain.
MultiWave: Multiresolution Deep Architectures through Wavelet Decomposition for Multivariate Time Series Prediction
information-fusion-lab-umass/multiwave
•
•
To address these issues, we introduce MultiWave, a novel framework that enhances deep learning time series models by incorporating components that operate at the intrinsic frequencies of signals.
Vision-Language Models can Identify Distracted Driver Behavior from Naturalistic Videos
Our results show that this framework offers state-of-the-art performance on zero-shot transfer and video-based CLIP for predicting the driver's state on two public datasets.
