Developing an explainable human action recognition system for academic environments: Enhancing educational interaction

• Explainable Action Recognition for Education : Enhances educational interactions in academia. • Temporal Segment Network (TSN) : Captures dynamic temporal patterns for robust action recognition. • Segmented Video Processing : Analyzes video segments independently for comprehensive action insights. • Improved Communication for Non-Verbal Cues : Valuable for individuals with communication challenges, highlighting gestures. • High Accuracy and Minimal Loss: achieves a high recognition accuracy of 98.6 % with minimal loss. This study presents the development of an Explainable Human Action Recognition System tailored for academic environments, aiming to enhance educational interactions. By leveraging a Temporal Segment Network as the core methodology, the system achieves robust action recognition through a multi-segment approach that captures dynamic temporal patterns in video data. The TSN divides video sequences into several segments, processes them individually, and integrates the information to provide a comprehensive understanding of the observed actions. This allows the system to discern and classify a wide range of educational activities, from classroom gestures to interactive learning behaviors. A critical feature of the proposed system is its explainability, ensuring that the recognized actions and the underlying decision-making processes are transparent and interpretable to educators and learners alike. Human action recognition is particularly valuable in contexts where non-verbal communication is prevalent, such as in interactions involving individuals with communication challenges, where gestures play a crucial role in conveying information and emotions. Our results demonstrate that this approach not only improves the accuracy of action recognition, achieving a 98.6 % accuracy with minimal loss, but also supports the detailed analysis of educational interactions, ultimately contributing to more engaging and effective learning experiences.

Authors

Institutions

Publication Details

Journal
Results in Engineering
Published
2025-04-25
DOI
https://doi.org/10.1016/j.rineng.2025.105014
Citations
4
Primary Topic
Human Pose and Action Recognition
Type
article
Field-Weighted Citation Impact
5.10
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Developing an explainable human action recognition system for academic environments: Enhancing educational interaction

Mustafa Basthikodi, Poornima B-V
4 citations
Results in Engineering
Human Pose and Action Recognition
5.10
article

Developing an explainable human action recognition system for academic environments: Enhancing educational interaction

Mustafa Basthikodi, Poornima B-V
article en
4 citations

Abstract

• Explainable Action Recognition for Education : Enhances educational interactions in academia. • Temporal Segment Network (TSN) : Captures dynamic temporal patterns for robust action recognition. • Segmented Video Processing : Analyzes video segments independently for comprehensive action insights. • Improved Communication for Non-Verbal Cues : Valuable for individuals with communication challenges, highlighting gestures. • High Accuracy and Minimal Loss: achieves a high recognition accuracy of 98.6 % with minimal loss. This study presents the development of an Explainable Human Action Recognition System tailored for academic environments, aiming to enhance educational interactions. By leveraging a Temporal Segment Network as the core methodology, the system achieves robust action recognition through a multi-segment approach that captures dynamic temporal patterns in video data. The TSN divides video sequences into several segments, processes them individually, and integrates the information to provide a comprehensive understanding of the observed actions. This allows the system to discern and classify a wide range of educational activities, from classroom gestures to interactive learning behaviors. A critical feature of the proposed system is its explainability, ensuring that the recognized actions and the underlying decision-making processes are transparent and interpretable to educators and learners alike. Human action recognition is particularly valuable in contexts where non-verbal communication is prevalent, such as in interactions involving individuals with communication challenges, where gestures play a crucial role in conveying information and emotions. Our results demonstrate that this approach not only improves the accuracy of action recognition, achieving a 98.6 % accuracy with minimal loss, but also supports the detailed analysis of educational interactions, ultimately contributing to more engaging and effective learning experiences.

Results in EngineeringVol. 26
Sahyadri Hospital (IN)
Openalex Percentile: Top 5%
Human Pose and Action Recognition
5.10
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.