Synthetic Training Data Generation for Activity Monitoring and Behavior Analysis

Monekosso, Dorothy; Remagnino, Paolo

doi:10.1007/978-3-642-05408-2_31

Cited by 5 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We would like to adapt our approaches to handle such scenarios and by benefiting from the unusual event data as they are always scarce. To model specific abnormal activities which have either very few data to begin with or data collection is difficult, approaches based on over-sampling of minority class [3], and synthetic data generation for activity recognition [20,19] can be explored to simulate more data and generative classification approaches be used. In such cases, investigating the role of discriminative temporal classifiers such as Conditional Random Fields [28,16] will be interesting to explore.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

Towards the detection of unusual temporal events during activities using HMMs

Khan

Karg

Hoey

et al. 2012

Proceedings of the 2012 ACM Conference on Ubiquitous Computing

View full text Add to dashboard Cite

Most of the systems for recognition of activities aim to identify a set of normal human activities. Data is either recorded by computer vision or sensor based networks. These systems may not work properly if an unusual event or abnormal activity occurs, especially ones that have not been encountered in the past. By definition, unusual events are mostly rare and unexpected, and therefore very little or no data may be available for training. In this paper, we focus on the challenging problem of detecting unusual temporal events in a sensor network and present three Hidden Markov Models (HMM) based approaches to tackle this problem. The first approach models each normal activity separately as an HMM and the second approach models all the normal activities together as one common HMM. If the likelihood is lower than a threshold, an unusual event is identified. The third approach models all normal activities together in one HMM and approximates an HMM for the the unusual events. All the methods train HMM models on data of the usual events and do not require training data from the unusual events. We perform our experiments on a Locomotion Analysis dataset that contains gyroscope, force sensor, and accelerometer readings. To test the performance of our approaches, we generate five types of unusual events that represent random activity, extremely unusual events, unusual events similar to specific normal activities, no or little motion and normal activity followed by no or little motion. Our experiments suggest that for a moderately sized time frame window, these approaches can identify all the five types of unusual events with high confidence.

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

Towards the detection of unusual temporal events during activities using HMMs

Khan

Karg

Hoey

et al. 2012

Proceedings of the 2012 ACM Conference on Ubiquitous Computing

View full text Add to dashboard Cite

show abstract

“…Most publicly available activity datasets are small for various reasons such as unavailability of participants and need for data collection over long periods of time (Mendez-Vazquez, Helal, & Cook, 2009). However, evaluation of privacy-preserving approaches typically requires large data sets to achieve desired levels of accuracy (Kitamura, Chen, & Pendyala, 1997;Monekosso & Remagnino, 2009). To address this problem, a larger synthetic data set was generated from Student Life dataset (Wang et al, 2014) by paying attention to preserving the distributional characteristics of the original data.…”

Section: Datamentioning

confidence: 99%

An Effective and Computationally Efficient Approach for Anonymizing Large-Scale Physical Activity Data

Parameshwarappa

Chen

Koru

2021

Research Anthology on Privatizing and Securing Data

View full text Add to dashboard Cite

Publishing physical activity data can facilitate reproducible health-care research in several areas such as population health management, behavioral health research, and management of chronic health problems. However, publishing such data also brings high privacy risks related to re-identification which makes anonymization necessary. One of the challenges in anonymizing physical activity data collected periodically is its sequential nature. The existing anonymization techniques work sufficiently for cross-sectional data but have high computational costs when applied directly to sequential data. This article presents an effective anonymization approach, multi-level clustering-based anonymization to anonymize physical activity data. Compared with the conventional methods, the proposed approach improves time complexity by reducing the clustering time drastically. While doing so, it preserves the utility as much as the conventional approaches.

show abstract

“…Traditional data collection, (e.g., bootstrapping) methods may affect the size and quality of the generated data sets [5]. In this paper, novel approaches to generate replicates (or extensions) of the datasets are proposed and discussed.…”

Section: Introductionmentioning

confidence: 99%

Statistical vs. visual data generation in hand gesture recognition

Ibrahim

Kashef

2012

2012 Seventh International Conference on Computer Engineering &Amp; Systems (ICCES)

View full text Add to dashboard Cite

A dataset with diverse training data is essence of the hand gesture recognition research. Most of the benchmarked datasets are limited in the number of signers and/or the number of each gesture try, which often result in over-fitting and poor generalization. Overcoming this challenge is often achieved by collecting a large number of exemplars for each hand gesture. This process is either expensive or impractical. Recently, synthetic data generation methods have been presented as a more reliable way to extend and enrich datasets. This paper proposes a comparative study of statistical and visual synthetic data generation. The visual synthetic data generation is executed by building synthetic 3D animated models using human figure design software. The experiments illustrate how the recognition accuracy will be changed when both methods used in enlarging the training data. The results show that in both cases recognition accuracy is enhanced, and in most cases, the visual synthetic data enlargement provides better improvement in the quality and diversity of the training data.

show abstract

Synthetic Training Data Generation for Activity Monitoring and Behavior Analysis

Cited by 5 publications

References 12 publications

Towards the detection of unusual temporal events during activities using HMMs

Towards the detection of unusual temporal events during activities using HMMs

An Effective and Computationally Efficient Approach for Anonymizing Large-Scale Physical Activity Data

Statistical vs. visual data generation in hand gesture recognition

Contact Info

Product

Resources

About