Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ARIC: An Activity Recognition Dataset in Classroom Surveillance Images

Linfeng Xu, Fanman Meng, Qingbo Wu, Lili Pan, Heqian Qiu, Lanxiao Wang, Kailong Chen, Kanglei Geng, Yilei Qian, Haojie Wang, Shuchang Zhou, Shimou Ling, Zejia Liu, Nanlin Chen, Yingjie Xu, Shaoxu Cheng, Bowen Tan, Ziyong Xu, Hongliang Li

TL;DR

The paper tackles the gap in publicly available, multimodal activity recognition data for real classroom surveillance while addressing privacy concerns. It introduces ARIC, a dataset with 36,453 surveillance images across 32 activities, plus audio and text modalities, captured from real classrooms with multi-angle views. Privacy-preserving releases provide shallow features from pretrained models (ResNet50, ViT, CLIP-ViT) and a supplementary ARIC_supplement with annotations, enabling safer dissemination and analysis. It also formalizes continual learning and few-shot continual learning scenarios to mimic open teaching environments, offering standardized settings and resources to spur future research in classroom AI and education technology.

Abstract

The application of activity recognition in the ``AI + Education" field is gaining increasing attention. However, current work mainly focuses on the recognition of activities in manually captured videos and a limited number of activity types, with little attention given to recognizing activities in surveillance images from real classrooms. Activity recognition in classroom surveillance images faces multiple challenges, such as class imbalance and high activity similarity. To address this gap, we constructed a novel multimodal dataset focused on classroom surveillance image activity recognition called ARIC (Activity Recognition In Classroom). The ARIC dataset has advantages of multiple perspectives, 32 activity categories, three modalities, and real-world classroom scenarios. In addition to the general activity recognition tasks, we also provide settings for continual learning and few-shot continual learning. We hope that the ARIC dataset can act as a facilitator for future analysis and research for open teaching scenarios. You can download preliminary data from https://ivipclab.github.io/publication_ARIC/ARIC.

ARIC: An Activity Recognition Dataset in Classroom Surveillance Images

TL;DR

The paper tackles the gap in publicly available, multimodal activity recognition data for real classroom surveillance while addressing privacy concerns. It introduces ARIC, a dataset with 36,453 surveillance images across 32 activities, plus audio and text modalities, captured from real classrooms with multi-angle views. Privacy-preserving releases provide shallow features from pretrained models (ResNet50, ViT, CLIP-ViT) and a supplementary ARIC_supplement with annotations, enabling safer dissemination and analysis. It also formalizes continual learning and few-shot continual learning scenarios to mimic open teaching environments, offering standardized settings and resources to spur future research in classroom AI and education technology.

Abstract

The application of activity recognition in the ``AI + Education" field is gaining increasing attention. However, current work mainly focuses on the recognition of activities in manually captured videos and a limited number of activity types, with little attention given to recognizing activities in surveillance images from real classrooms. Activity recognition in classroom surveillance images faces multiple challenges, such as class imbalance and high activity similarity. To address this gap, we constructed a novel multimodal dataset focused on classroom surveillance image activity recognition called ARIC (Activity Recognition In Classroom). The ARIC dataset has advantages of multiple perspectives, 32 activity categories, three modalities, and real-world classroom scenarios. In addition to the general activity recognition tasks, we also provide settings for continual learning and few-shot continual learning. We hope that the ARIC dataset can act as a facilitator for future analysis and research for open teaching scenarios. You can download preliminary data from https://ivipclab.github.io/publication_ARIC/ARIC.

Paper Structure

This paper contains 8 sections, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. ARIC-Dataset
  3. Data Collection
  4. Dataset Overview
  5. Privacy Protection
  6. Settings for Continual Learning
  7. Settings for Few-Shot Continual Learning
  8. The Supplement of ARIC

Figures (4)

  • Figure 1: Monitoring Images from Different Perspectives
  • Figure 2: Sample Distribution of the 32 Activity Categories
  • Figure 3: A Setting for Continual Learning Setting
  • Figure 4: A Setting for Few-Shot Continual Learning