Feedback based real time facial and head gesture recognition for e-learning system

Deshmukh, Shubhada P.; Patwardhan, Manasi; Mahajan, Annil

doi:10.1145/3152494.3167991

Cited by 7 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the rapid development and lower cost of wearable devices with embedded sensors, a plethora of new applications, such as healthcare [1], authentication [2], robotic control [3], virtual/augmented reality [4], [5], [6] and e-learning [7] are emerging. However, a number of unique challenges need to be addressed in order to best harness the promise of such applications and enable the widespread use of wearable technology.…”

Section: Introductionmentioning

confidence: 99%

Conditional-UNet: A Condition-aware Deep Model for Coherent Human Activity Recognition From Wearables

Zhang

Japkowicz

2021

2020 25th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

Recognizing human activities from multi-channel time series data collected from wearable sensors has become an important practical application of machine learning. A serious challenge comes from the presence of coherent activities or body movements, such as movements of the head while walking or sitting, since signals representing these movements are mixed and interfere with each other. Basic multi-label classification typically assumes independence within the multiple activities. This is oversimplified and reduces modeling power even when using stateof-the-art deep learning methods. In this paper, we investigate this new problem, which we name "Coherent Human Activity Recognition (Co-HAR)", that keeps complete conditional dependency between the multiple labels. Additionally, we treat Co-HAR as a dense labelling problem that classifies each sample on a time step with multiple coherent labels to provide high-fidelity and duration-sensitive support to high-precision applications. To explicitly model conditional dependency, a novel conditionaware deep architecture "Conditional-UNet" is developed to allow for multiple dense labeling for Co-HAR. We also contribute a first-of-its-kind Co-HAR dataset for head gesture recognition associated with a user's activity, walking or sitting, to the research community. Extensive experiments on this dataset show that our model outperforms state-of-the-art deep learning methods and achieves up to 92% accuracy on context-based head gesture classification.

show abstract

Section: Introductionmentioning

confidence: 99%

Conditional-UNet: A Condition-aware Deep Model for Coherent Human Activity Recognition From Wearables

Zhang

Japkowicz

2021

2020 25th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

show abstract

The Evidence of Impact and Ethical Considerations of Multimodal Learning Analytics: A Systematic Literature Review

Alwahaby

Cukurova

Papamitsiou

et al. 2022

The Multimodal Learning Analytics Handbook

View full text Add to dashboard Cite

There is a growing interest in the research and use of multimodal data in learning analytics. This paper presents a systematic literature review of multimodal learning analytics (MMLA) research to assess i) the available evidence of impact on learning outcomes in real-world contexts and ii) explore the extent to which ethical considerations are addressed. A few recent literature reviews argue for the promising value of multimodal data in learning analytics research. However, our understanding of the challenges associated with MMLA research from real-world teaching and learning environments is limited. To address this gap, this paper provides an overview of the evidence of impact and ethical considerations stemming from an analysis of the relevant MMLA research published in the last decade. The search of the literature resulted in 663 papers, of which 100 were included in the final synthesis. The results show that the evidence of real-world impact on learning outcomes is weak, and ethical aspects of MMLA work are rarely addressed. We discuss our results through the lenses of two theoretical frameworks for evidence of impact types and ethical dimensions of MMLA. We conclude that for MMLA to stay relevant and become part of mainstream education, future research should directly address the gaps identified in this review.

show abstract

Multimodal Emotion Analytics for E-Learning

Devi

Reddy

2021

Intelligent Data Communication Technologies and Internet of Things

View full text Add to dashboard Cite

Feedback based real time facial and head gesture recognition for e-learning system

Cited by 7 publications

References 14 publications

Conditional-UNet: A Condition-aware Deep Model for Coherent Human Activity Recognition From Wearables

Conditional-UNet: A Condition-aware Deep Model for Coherent Human Activity Recognition From Wearables

The Evidence of Impact and Ethical Considerations of Multimodal Learning Analytics: A Systematic Literature Review

Multimodal Emotion Analytics for E-Learning

Contact Info

Product

Resources

About