CADDI: An in-Class Activity Detection Dataset using IMU data from low-cost sensors

2025-03-04

AI Summary

To address the scarcity of high-quality labeled data for student behavior recognition in classroom settings, this paper introduces the first open-source, multimodal IMU-based activity dataset specifically designed for educational environments. The dataset encompasses 19 instantaneous and continuous behaviors performed by 12 students in authentic classroom conditions, with synchronized, low-power IMU (accelerometer, gyroscope, rotation vector) and stereo visual data acquisition, followed by precise temporal alignment and fine-grained annotation. In contrast to prior work, our dataset uniquely unifies education-specific design, large-scale coverage, multimodal sensing, high temporal fidelity, and cost-effective deployability. It thus fills a critical gap in the field by providing a high-quality, standardized benchmark for classroom behavior recognition—enabling robust training and fair evaluation of activity recognition models with improved accuracy and generalization capability.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Video Understanding & Activity Analysis
• Machine Learning: Multimodal Learning

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

The monitoring and prediction of in-class student activities is of paramount importance for the comprehension of engagement and the enhancement of pedagogical efficacy. The accurate detection of these activities enables educators to modify their lessons in real time, thereby reducing negative emotional states and enhancing the overall learning experience. To this end, the use of non-intrusive devices, such as inertial measurement units (IMUs) embedded in smartwatches, represents a viable solution. The development of reliable predictive systems has been limited by the lack of large, labeled datasets in education. To bridge this gap, we present a novel dataset for in-class activity detection using affordable IMU sensors. The dataset comprises 19 diverse activities, both instantaneous and continuous, performed by 12 participants in typical classroom scenarios. It includes accelerometer, gyroscope, rotation vector data, and synchronized stereo images, offering a comprehensive resource for developing multimodal algorithms using sensor and visual data. This dataset represents a key step toward scalable solutions for activity recognition in educational settings.

Problem

• Detecting in-class student activities using IMU data
• Enhancing pedagogical efficacy through real-time activity monitoring
• Addressing lack of labeled datasets for educational activity recognition

Innovation

• Uses IMU data from low-cost smartwatch sensors
• Includes 19 diverse classroom activities dataset
• Combines sensor and visual data for multimodal algorithms