AI summary · arXiv: 2303.10916v1 · cs.CV

Learning Behavior Recognition in Smart Classroom with Multiple Students Based on YOLOv5

Zhifeng Wang, Jialong Yao, Chunyan Zeng, Wanxuan Wu, Hongmin Xu, Yang Yang
Original paper Submitted to arXiv on 20 Mar. 2023 8 pages, 10 figures License: NONEXCLUSIVE-DISTRIB 1.0

How to read this summary? Illustrated guide

The essentials

  • Computer vision technology is increasingly used to identify students' learning behavior in the classroom
  • Existing systems have limitations in accurately tracking and detecting multiple targets
  • A team of researchers proposed a YOLOv5s network structure based on the You Only Look Once (YOLO) algorithm to recognize and analyze students' classroom behavior
  • The proposed method involves pre-processing input images, extracting deep features through convolutional layers, applying Squeeze-and-Excitation (SE) attention detection mechanism, and using Feature Pyramid Networks (FPN) and Path Aggregation Network (PAN) structures to classify extracted features
  • Multiple experiments were conducted comparing traditional methods with the proposed method's effectiveness, showing an 11% improvement in mean Average Precision (mAP) performance compared to YOLOv4
  • The proposed method can significantly reduce traditional teachers' workload while ensuring greater accuracy and comprehensiveness in supervising students’ behaviors
  • Cross-fertilization using computer vision technology can enhance deep learning-based computer vision technology's strength further
  • Technological advancements can improve education by providing more efficient ways of monitoring students' behaviors in classrooms

Key points written by AI: they may contain errors, check them against the paper.

AI-generated keywords: Computer Vision YOLOv5s SE Attention FPN PAN

Read further

  • Comprehensive Summary
  • Layman's Summary
  • Blog article
  • Original abstract
The use of computer vision technology in identifying students' learning behavior in the classroom has become increasingly popular in recent years. However, existing student learning behavior detection systems have limitations in tracking and detecting multiple targets accurately, and their accuracy is not high enough to meet the needs for accurate recognition of student behavior. To address this issue, a team of researchers proposed a YOLOv5s network structure based on the You Only Look Once (YOLO) algorithm to recognize and analyze students' classroom behavior. The proposed method involves pre-processing input images taken in the smart classroom before feeding them into the designed YOLOv5 networks to extract deep features through convolutional layers. The Squeeze-and-Excitation (SE) attention detection mechanism is then applied to reduce the weight of background information during the recognition process. Finally, Feature Pyramid Networks (FPN) and Path Aggregation Network (PAN) structures are used to classify the extracted features. Multiple experiments were conducted to compare traditional learning behavior recognition methods with the proposed method's effectiveness. The results showed that compared with YOLOv4, the proposed method improved mean Average Precision (mAP) performance by 11%. This improvement can significantly reduce traditional teachers' workload while ensuring greater accuracy and comprehensiveness in supervising students’ behaviors. The study highlights how cross-fertilization using computer vision technology can enhance deep learning-based computer vision technology's strength further. It also demonstrates how technological advancements can improve education by providing more efficient ways of monitoring students' behaviors in classrooms.

Was this synthesis useful?

Score: 0

Why do we need votes?

Votes are used to determine whether we need to re-run our summarizing tools. If the count reaches -10, our tools can be restarted.

Figures and plots

Ask about a figure by number or use its question button to request visual analysis. Up to two figures and four image panels per question. Values read from plots are approximate; check the original data.

AI assistant instructions?

Tools for this paper

Explain it differently 1 free rewrite left today
Save to a collection Group papers in collections to compare them, chat with them and export them. Continue with Google Create a free account Log in

Advanced tools Premium

For researchers who follow a field closely: citation alerts, journal-club slides, version comparison and Zotero export. Discover Premium

Zotero Zotero guide
Citation alerts 13 papers already cite it: be emailed each new one. Citation alerts guide
Who works on it?

Among the 31 papers around this one, the name that comes back most: Zhifeng Wang (4 papers)

1 more person signs several of them. The full map adds their institutions and approaches, from the papers of the last years.

Continue with Google Create a free account: your first map is free
Weekly letter The latest letter of its field, "Artificial intelligence this week" of 16 November 2026: Artificial intelligence this week: What impressive scores leave out · Receive one by email for free

Papers around this one

See them on a map

Ranked by how close their title and abstract are to this paper’s. Select some to read them together.

Closest papers

  1. Very close Builds on this paper Cited by this paper
    YOLOv5 Enhanced Learning Behavior Recognition and Analysis in Smart Classroom with Multiple Students
    Zhifeng Wang, Jialong Yao, Chun-Yan Zeng et al. · 2022 · 2022 International Conference on Intelligent Education and Intelligent Research (IEIR) · cited 36 times · Not summarized yet

    YOLOv5 Enhanced Learning Behavior Recognition extends the same multiple-student smart-classroom behavior-recognition work and is cited by this paper.

    Abstract

    Deep learning-based computer vision technology has grown stronger in recent years, and cross-fertilization using computer vision technology has been a popular direction in recent years. The use of computer vision technology to identify students’ learning behavior in the classroom can reduce the workload of traditional teachers in supervising students in the classroom, and ensure greater accuracy and comprehensiveness. However, existing student learning behavior detection systems are unable to track and detect multiple targets precisely, and the accuracy of learning behavior recognition is not high enough to meet the existing needs for the accurate recognition of student behavior in the class…

  2. Very close Another approach to the problem
    A Classroom Behavior Recognition Method for Students based on YOLOv10
    Shuai Ma, Jian Han, Tong He et al. · 2025 · International Conference on Advanced Computer Theory and Engineering · cited 1 time · Not summarized yet

    The YOLOv10 method recognizes student classroom behaviors, addressing the same task with a later YOLO detector.

    Abstract

    With the widespread adoption of smart education concepts, the limitations of traditional large-classroom management methods have become increasingly apparent. The application of student classroom behavior recognition technology is highly beneficial in helping teachers improve the quality of classroom education and instruction. Currently, the dataset composition for student classroom behavior recognition remains incomplete, and challenges arise in recognizing overlapping targets and dense small objects. To address these issues, this paper designs a real-time analysis system for student classroom behavior based on YOLOv10. The specific work is as follows:First, to address the problem of imbala…

  3. Very close Another approach to the problem
    Student Classroom Behavior Recognition Based on Improved YOLOv7
    Chunling Li, Cheng Peng · 2025 · Proceedings of the 2025 2nd International Conference on Informatics Education and Computer Technology Applications · cited 3 times · Not summarized yet

    The improved YOLOv7 method recognizes student behavior in cluttered, crowded classrooms using CBAM attention modules.

    Abstract

    The recognition of students' classroom behavior by artificial intelligence and deep learning can be used as an auxiliary means to improve teaching quality. In the real classroom scene, due to the background clutter, crowded scene and object occlusion and other problems, There are many challenges in behavior recognition. To solve these problems, This paper proposes an improved model based on YOLOv7. It is used to identify students' classroom behavior in real classroom scenes. The improved model incorporates three CBAM attention modules into the original structure of YOLOv7. To obtain additional local information about the image during training and ensure the recognition accuracy. In order to …

  4. Very close Another approach to the problem
    Student classroom behavior detection based on deep learning
    Xiaochen Yang, Ke Hu · 2025 · International Conference on Image Processing, Intelligent Control and Computer Engineering · cited 2 times · Not summarized yet

    The YOLO-SE model detects four student behaviors with YOLOv8, rather than this paper’s YOLOv5s approach.

    Abstract

    With the rapid development of artificial intelligence, student classroom behavior detection based on computer vision has provided important support for the intelligentization of education. Traditional classroom behavior recognition relies heavily on manual observation and video playback, which has problems such as low efficiency and strong subjectivity. This study proposes an improved student classroom behavior detection model YOLO-SE based on YOLOv8, aiming to achieve accurate recognition of four types of behaviors: raising hand, reading, dozing off, and writing, and to enhance the model's robustness in complex scenarios. Firstly, the feature extraction ability is optimized by introducing t…

  5. Very close Another approach to the problem
    Classroom Behavior Recognition of Students Based on YOLOv8-OGCA
    Hong-Qian Chen, Yu Gao, Rui Yang · 2024 · ICAIE · cited 1 time · Not summarized yet

    YOLOv8-OGCA recognizes classroom student behavior using coordinate attention and SPD-Conv to address small targets.

    Abstract

    Behavior detection is a task in computer vision, aiming to promote teachers' professionalism and increase classroom quality by studying students' behavior. Using deep learning methods to automatically detect students' behavior in the classroom can effectively improve the efficiency of analyzing students' classroom performance. This article proposes a method for recognizing students' classroom behavior based on YOLOv8n: YOLOv8-OGCA. First, the module global grouping coordinate attention OGCA is proposed. Then, SPD-Conv is introduced to address the problem of small detection targets for each student in student behavior images, replacing some cross-convolution layers and pooling layers to reduc…

  6. Very close Another approach to the problem
    Student Behavior Recognition in Classroom Based on Deep Learning
    Qingzheng Jia, Jialiang He · 2024 · Applied Sciences · cited 55 times · Not summarized yet

    The SBD model combines YOLOv5, contextual attention, and OpenPose for student behavior recognition in complex classrooms.

    Abstract

    With the widespread application of information technology in education, the real-time detection of student behavior in the classroom has become a key issue in improving teaching quality. This paper proposes a Student Behavior Detection (SBD) model that combines YOLOv5, the Contextual Attention (CA) mechanism and OpenPose, aiming to achieve efficient and accurate behavior recognition in complex classroom environments. By integrating YOLOv5 with the CA attention mechanism to enhance feature extraction capabilities, the model’s recognition performance in complex backgrounds, such as those with occlusion, is significantly improved. In addition, the feature map generated by the improved YOLOv5 is…

  7. Very close Another approach to the problem
    Intelligent Classroom Student Behavior Recognition Based on Enhanced YOLOv8
    Xinhui Wang, Yuan-Yuan Gao, Yun-Fei Zhang et al. · 2025 · ICAIE · cited 1 time · Not summarized yet

    The enhanced YOLOv8 framework recognizes classroom behaviors using multi-scale feature fusion and attention.

    Abstract

    Accurate recognition of student classroom behaviors in intelligent teaching environments is crucial for improving teaching quality and learning effectiveness. To address technical challenges in real classroom scenarios, such as large-scale target variations, severe occlusions, and dense distributions, this paper proposes an enhanced YOLOv8-based student behavior recognition framework. First, by integrating the multi-scale feature extraction capability of Res2Net with the efficient detection structure of YOLOv8, we design a Multi-Scale Feature Fusion Block (MSF-Block) to enhance multi-scale feature learning capabilities. Second, we integrate an Efficient Channel-Spatial Attention mechanism (E…

  8. Very close Another approach to the problem
    Research and Application of Student Classroom Behavior Recognition Based on Deep Learning
    Jian-Xia Wang, Dong-Min Huo, Chunjin Yang et al. · 2024 · 2024 6th Asia Symposium on Image Processing (ASIP) · cited 1 time · Not summarized yet

    The improved YOLOv8 model addresses classroom student behavior detection with Alpha-SIOU loss and multidimensional collaborative attention.

    Abstract

    With the rapid development of artificial intelligence technology, computer vision technology has become an important means of obtaining behavioral information. However, the current student classroom behavior detection system still has the problem of low target detection accuracy. To improve students' learning efficiency in the classroom and reduce teachers' workload, this article proposes an improved YOLOv8 network model. The bounding box loss function is modified to Alpha-SIOU and the multi-dimensional collaborative attention mechanism MCA is introduced in front of the head to optimize the model. Experimental results show that when the detection rate and calculation amount of this model rem…

Prior work it builds on

  1. Close Broader context
    Automation of Recording in Smart Classrooms via Deep Learning and Bayesian Maximum a Posteriori Estimation of Instructor's Pose
    M. S. Haghighi, Alireza Sheikhjafari, A. Jolfaei et al. · 2021 · IEEE Transactions on Industrial Informatics · cited 15 times · Not summarized yet

    The camera-management system uses deep learning to detect an instructor’s position and pose for recording smart-classroom lectures.

    Abstract

    Internet of Things is making objects smarter and more autonomous. At the other side, online education is gaining momentum and many universities are now offering online degrees. Content preparation for such programs usually involves recording the classes. In this article, we intend to introduce a deep learning-based camera management system as a substitute for the academic filming crew. The solution mainly consists of two cameras and a wearable gadget for the instructor. The fixed camera is used for the instructor's position and pose detection and the pan–tilt–zoom (PTZ) camera does the filming. In the proposed solution, image processing and deep learning techniques are merged together. Face …

  2. Related Same method, another problem A method it uses
    Ratio-and-Scale-Aware YOLO for Pedestrian Detection
    Wei-Yen Hsu, Wenbin Lin · 2020 · IEEE Transactions on Image Processing · cited 175 times · Not summarized yet

    RSA-YOLO modifies a YOLO detector for pedestrian detection, applying the same detection-model family to a different object.

    Abstract

    Current deep learning methods seldom consider the effects of small pedestrian ratios and considerable differences in the aspect ratio of input images, which results in low pedestrian detection performance. This study proposes the ratio-and-scale-aware YOLO (RSA-YOLO) method to solve the aforementioned problems. The following procedure is adopted in this method. First, ratio-aware mechanisms are introduced to dynamically adjust the input layer length and width hyperparameters of YOLOv3, thereby solving the problem of considerable differences in the aspect ratio. Second, intelligent splits are used to automatically and appropriately divide the original images into two local images. Ratio-aware…

  3. Related Broader context
    Joint Motion Information Extraction and Human Behavior Recognition in Video Based on Deep Learning
    Kai Zhang, Wen-Jie Ling · 2020 · IEEE Sensors Journal · cited 18 times · Not summarized yet

    The video-recognition study extracts joint motion features with two-channel deep networks to recognize human behavior.

    Abstract

    Video surveillance has played a huge role in the safety and security tasks of various places for many years, and its application range is also expanding. The core problem that needs to be addressed in intelligent video surveillance is the identification and analysis of human behavior in video. In this paper, the structure and joint motion feature extraction of the two-channel deep convolutional neural network model are studied and designed. In the design of the network structure, the ventral and dorsal channels used in the processing of visual signals by the brain visual cortex were simulated. The spatial channel network and the time channel network are used to process static information and…

  4. Related Same method, another problem
    Real-Time Vehicle Detection Based on Improved YOLO v5
    Yu Zhang, Zhong-Yin Guo, Jian-Qing Wu et al. · 2022 · Sustainability · cited 462 times · Not summarized yet

    The vehicle detector improves YOLOv5 and uses image augmentation, applying the same base detector to traffic scenes.

    Abstract

    To reduce the false detection rate of vehicle targets caused by occlusion, an improved method of vehicle detection in different traffic scenarios based on an improved YOLO v5 network is proposed. The proposed method uses the Flip-Mosaic algorithm to enhance the network’s perception of small targets. A multi-type vehicle target dataset collected in different scenarios was set up. The detection model was trained based on the dataset. The experimental results showed that the Flip-Mosaic data enhancement algorithm can improve the accuracy of vehicle detection and reduce the false detection rate.

  5. Related Same method, another problem
    Face Mask Recognition System with YOLOV5 Based on Image Recognition
    Guanhao Yang, Wei Feng, Jintao Jin et al. · 2020 · 2020 IEEE 6th International Conference on Computer and Communications (ICCC) · cited 177 times · Not summarized yet

    The mask-recognition system uses YOLOv5 for face-mask detection, applying the same model family to another recognition task.

    Abstract

    The rapid development of computer vision makes human-computer interaction possible and has a wide application prospect. Since the discovery of the first case of COVID-19, the global fight against the epidemic has begun. In addition to various studies and findings by medical and health care experts, people's daily behaviors have also become key to combating the epidemic. In China, the government has taken active and effective measures of isolation and closure, as well as the active cooperation of the general public, such as it is unnecessary to stay indoors and wear masks. China, as the country with the first outbreak of the epidemic, has now become the benchmark country of epidemic preventio…

Papers that cite it

  1. Very close Builds on this paper
    Improving YOLOv7 for Large Target Classroom Behavior Recognition of Teachers in Smart Classroom Scenarios
    Long Ma, Tao Zhou, Baohua Yu et al. · 2024 · Electronics · cited 18 times · Not summarized yet

    The teacher-behavior study improves YOLOv7 for smart classrooms and cites this paper as prior classroom behavior work.

    Abstract

    Deep learning technology has recently become increasingly prevalent in the field of education due to the rapid growth of artificial intelligence. Teachers’ teaching behavior is a crucial component of classroom teaching activities, and identifying and examining teachers’ classroom teaching behavior is an important way to assess teaching. However, the traditional teaching evaluation method involves evaluating by either listening to the class on-site or playing back the teaching video afterward, which is a time-consuming and inefficient manual method. Therefore, this paper obtained teaching behavior data from a real smart classroom scenario and observed and analyzed the teacher behavior charact…

  2. Very close Builds on this paper
    Object detection of classroom students based on improved YOLOv7
    Hui-Bai Wang, Xin-Xin Chen · 2024 · Other Conferences · cited 1 time · Not summarized yet

    The YOLOv7 student-detection study addresses classroom occlusion and dense targets, and cites this paper.

    Abstract

    With the advancement and maturity of computer vision technology, its application in the field of education has become a new trend. This paper presents a research study on the improvement of YOLOv7 for classroom student object detection, which holds significant implications for integrating computer vision technology into the educational domain. During the detection of classroom student targets using YOLOv7, challenges such as dense distribution, severe occlusion, and small target sizes were identified. These challenges led to issues like missed detections and false positives[1]. Addressing these issues, this paper introduces a repulsion loss function, building upon the original loss function …

  3. Very close Builds on this paper Uses this paper in depth
    HARM-YOLO11n: A Lightweight Real-Time Object Detection Framework for Robust Classroom Behavior Monitoring
    Zihang Zhang, Yonghua Mao, Ying-Cang Ma et al. · 2026 · Italian National Conference on Sensors · Not summarized yet

    HARM-YOLO11n develops a later YOLO framework for classroom behavior monitoring and cites this paper.

    Abstract

    Real-time classroom behavior monitoring is an important component of intelligent education systems, yet it remains challenging due to dense student layouts, frequent occlusions, and small-scale behavior targets. To address these challenges, this paper proposes Hierarchical Adaptive Re-parameterized Multi-scale YOLO11n (HARM-YOLO11n), a lightweight object detection framework based on YOLO11n for classroom behavior analysis. The proposed framework integrates structural re-parameterized feature enhancement (SRFE), a spatial-channel adaptive module (SCAM), position-adaptive multi-scale fusion (PAMSF), and IoU-adaptive soft weight (IASW) to improve feature representation, multi-scale interaction,…

  4. Very close Builds on this paper
    A Study of Classroom Behavior Recognition Incorporating Super-Resolution and Target Detection
    Xiaoli Zhang, Jialei Nie, Shou-Lin Wei et al. · 2024 · Italian National Conference on Sensors · cited 19 times · Not summarized yet

    The super-resolution study combines image enhancement with classroom behavior detection and cites this paper.

    Abstract

    With the development of educational technology, machine learning and deep learning provide technical support for traditional classroom observation assessment. However, in real classroom scenarios, the technique faces challenges such as lack of clarity of raw images, complexity of datasets, multi-target detection errors, and complexity of character interactions. Based on the above problems, a student classroom behavior recognition network incorporating super-resolution and target detection is proposed. To cope with the problem of unclear original images in the classroom scenario, SRGAN (Super Resolution Generative Adversarial Network for Images) is used to improve the image resolution and thu…

  5. Very close Builds on this paper
    Intelligent recognition of students’ behavior for smart learning environments
    Amr Abozeid, I. Alrashdi, R. M. Al-Makhlasawy · 2026 · Scientific Reports · cited 6 times · Not summarized yet

    The meta-learning study develops student behavior recognition for smart learning environments and cites this paper.

    Abstract

    The automatic detection of student behaviors is essential for improving smart classroom technologies and offering data-driven insights regarding student engagement. Nevertheless, existing methods encounter considerable obstacles caused by class imbalance, restricted annotations, and the slight visual resemblances among behavior categories. To overcome these constraints, we present a meta-learning framework that combines Vision Transformers with Prototypical Networks, improved by supervised contrastive learning and hard negative mining. The process starts by preprocessing and cropping the input images, utilizing YAML annotations to focus on behavior-specific areas. Every input is converted in…

How are these papers found?

We compare the title and abstract of this paper with those of every paper summarized on this site, and with the papers Semantic Scholar recommends, those it cites and those citing it (for a much-cited paper, its most recent citations and, from OpenAlex, the most cited ones). Among the prior work and the follow-ups, the papers that use each other in depth, a method it uses and the much-cited follow-ups come a little higher. The levels (very close, close, related) come from the similarity of the texts, not from a judgment of quality; * marks a paper without an abstract, compared by its title. The sentence under a paper is written by an AI from the two abstracts: check it in the papers. A paper marked not relevant by three readers leaves the list.

Updated on 6 November 2026.

The illustrated guide to the papers around a paper

The previous summary was created more than a year ago and can be re-run (if necessary) by clicking on the Run button below.

Disclaimer: The AI-based summarization tool and virtual assistant provided on this website may not always provide accurate and complete summaries or responses. We encourage you to carefully review and evaluate the generated content to ensure its quality and relevance to your needs.