The SPHERE Challenge: Activity Recognition with Multimodal Sensor Data

@article{Twomey2016TheSC,
  title={The SPHERE Challenge: Activity Recognition with Multimodal Sensor Data},
  author={Niall Twomey and Tom Diethe and Meelis Kull and Hao Song and M. Camplani and S. Hannuna and Xenofon Fafoutis and Ni Zhu and Przemyslaw Woznowski and Peter A. Flach and I. Craddock},
  journal={ArXiv},
  year={2016},
  volume={abs/1603.00797}
}
This paper outlines the Sensor Platform for HEalthcare in Residential Environment (SPHERE) project and details the SPHERE challenge that will take place in conjunction with European Conference on Machine Learning and Principles and Practice of Knowledge Discov- ery (ECML-PKDD) between March and July 2016. The SPHERE chal- lenge is an activity recognition competition where predictions are made from video, accelerometer and environmental sensors. Monitory prizes will be awarded to the top three… Expand
STARS: Soft Multi-Task Learning for Activity Recognition from Multi-Modal Sensor Data
TLDR
A soft probabilistic classification model for human activity recognition from multi-modal sensors in a smart home environment that employs a softmax multi-task learning approach to fit a joint model for all the rooms in the smart home, taking into account the diverse types of sensors available in different rooms. Expand
Probabilistic Sensor Fusion for Ambient Assisted Living
TLDR
Bayesian models for sensor fusion are introduced, which aims to address the challenges of fusion of heterogeneous sensor modalities, and how the two separate tasks of location prediction and activity recognition can be fused into a single model, which allows for simultaneous learning an prediction for both tasks. Expand
Location based Hierarchical Approach for Activity Recognition with Multi-modal Sensor Data
Accurate identification of human activities could help us provide a better patient recovery training guidance, or an early alarm of emergency that may happen to elder people, such as stroke, falls,Expand
Designing a video monitoring system for AAL applications: the SPHERE case study
IT based Healthcare platforms have been widely recognized by research communities and institutions as key players in the future of home-based health monitoring and care. Features like personalisedExpand
Online Feature Selection for Activity Recognition using Reinforcement Learning with Multiple Feedback
TLDR
A new methodology to perform feature selection on the device based on Reinforcement Learning to find the optimum balance between power consumption and accuracy, and extends the RL algorithm to address multiple sources of feedback, and uses them to tailor the policy in conjunction with estimating the feedback accuracy. Expand
Collection of a Diverse, Realistic and Annotated Dataset for Wearable Activity Recognition
TLDR
An analysis of data quality, through outlier detection and assesses the suitability of the dataset for the creation and validation of Activity Recognition models, finds a Random Forest provided the best classification. Expand
Dominance of Deep LSTM in Smartphone Sensor based Human Activity Classification
Human activity categorization is one of the challenging topics on which several types of researches have been conducted. With the improvement of modern technologies and algorithms, there have beenExpand
Activity Recognition from Multi-modal Sensor Data Using a Deep Convolutional Neural Network
TLDR
A Deep Convolutional Neural Network is proposed to perform the activity recognition task using the multi-modal data collected from a smart residential home to demonstrate the challenges of dealing with multi- modal data and highlight the importance of having a suitable number of samples within each class for sufficiently training and testing deep learning models. Expand
Classifying Imbalanced Multi-modal Sensor Data for Human Activity Recognition in a Smart Home using Deep Learning
TLDR
Whether fusing real-world imbalanced multi-modal sensor data improves classification results as opposed to using unimodal data is examined; and deep learning approaches to dealing with imbalancedmulti-modAL sensor data when using various resampling methods andDeep learning models are compared. Expand
Human Daily Activity Recognition for Healthcare Using Wearable and Visual Sensing Data
TLDR
An activity recognition system is introduced, which learns a nonlinear SVM algorithm to identify 20 different human activities from accelerometer and RGB-D camera data and early experimental results show that the proposed approach is promising and effective. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 12 REFERENCES
Active transfer learning for activity recognition
TLDR
A hierarchical Bayesian model is derived that is a natural fit to activity recognition from accelerometers and can achieve faster learning with fewer labels on a target domain than by either active or transfer learning alone. Expand
Bridging e-Health and the Internet of Things: The SPHERE Project
TLDR
An overview of this rapidly growing body of work on sensing systems in the home, as well as the implications for machine learning are presented, with an aim of uncovering the gap between the state of the art and the broad needs of healthcare services in ambient assisted living. Expand
Unsupervised learning of sensor topologies for improving activity recognition in smart environments
TLDR
The hypothesis is that learning patterns based on combinations of sensors will be more powerful than single sensors alone and this work shows that through the application of signal processing and information-theoretic techniques it can achieve classification performance better than that of the exhaustive approach, whilst only incurring a small cost in terms of computational resources. Expand
A multi-modal sensor infrastructure for healthcare in a residential environment
TLDR
A multi-modal system architecture for AAL remote healthcare monitoring in the home, gathering information from multiple, diverse (sensor) data sources is proposed. Expand
A comparative home activity monitoring study using visual and inertial sensors
  • L. Tao, T. Burghardt, +5 authors I. Craddock
  • Computer Science
  • 2015 17th International Conference on E-health Networking, Application & Services (HealthCom)
  • 2015
TLDR
This paper introduces the challenging SPHERE-H130 action dataset, and reports automatic recognition results at maximal temporal resolution, which indicate that a vision-based approach outperforms accelerometer provided by two phone-based inertial sensors by an average of 14.85% accuracy for home actions. Expand
Context modulation of Sensor Data Applied to Activity Recognition in Smart Homes
TLDR
It is shown that sensors in smart homes can dramatically adapt their sensor network topology and that this approach can be used to help understand various aspects of such sensor environments. Expand
An RSSI-based wall prediction model for residential floor map construction
TLDR
This paper focuses on using RF signals to predict the number of walls between a wearable device and an access point, and constructs wall prediction models that yield up to 91% identification rate. Expand
Bayesian Active Learning with Evidence-Based Instance Selection
TLDR
This work introduces a new active learning method that is based on on Bayesian model selection, and hence fits more concomitantly with the Bayesian framework than previous decision theoretic approaches, and is able to cope with situations that the simple but naive method of uncertainty sampling cannot. Expand
Smart-Homes for eHealth: Uncertainty Management and Calibration
We introduce the SPHERE project which is devoted to eHealth in a smart-home context, and discuss the implications for such a system in terms of the quantification and management of uncertainty forExpand
A comparative study of pose representation and dynamics modelling for online motion quality assessment
TLDR
The continuous- state HMM, combined with pose representation based on body-joints' location, outperforms standard discrete-state HMM approaches and other skeleton-based features in detecting gait abnormalities, as well as assessing deviations from the motion model on a frame-by-frame basis. Expand
...
1
2
...