Cunling Bian

Dr. Cunling Bian

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
9 February 2021 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
CBCunling Bian
Cunling Bian
FFFeng Feng
Feng Feng
LWLiang Wan
Liang Wan
SWSong Wang
Song Wang
Skeleton data have been extensively used for action recognition since they can robustly accommodate dynamic circumstances and complex backgrounds. To guarantee the action-recognition performance, we prefer to use advanced and time-consuming algorithms to get more accurate and complete skeletons from the scene. However, this may not be acceptable in time- and resource-stringent applications. In this paper, we explore the feasibility of using low-quality skeletons, which can be quickly and easily estimated from the scene, for action recognition. While the use of low-quality skeletons will surely lead to degraded action-recognition accuracy, in this paper we propose a structural knowledge distillation scheme to minimize this accuracy degradations and improve recognition model's robustness to uncontrollable skeleton corruptions. More specifically, a teacher which observes high-quality skeletons obtained from a scene is used to help train a student which only sees low-quality skeletons generated from the same scene. At inference time, only the student network is deployed for processing low-quality skeletons. In the proposed network, a graph matching loss is proposed to distill the graph structural knowledge at an intermediate representation level. We also propose a new gradient revision strategy to seek a balance between mimicking the teacher model and directly improving the student model's accuracy. Experiments are conducted on Kenetics400, NTU RGB+D and Penn action recognition datasets and the comparison results demonstrate the effectiveness of our scheme.

30 September 2025 in Humanities and Social Sciences Communications
Verified
Journal Article
Open access
PDF file
CBCunling Bian
Cunling Bian
XWXiaofang Wang
Xiaofang Wang
YHYingxue Huang
Yingxue Huang
SZShan Zhou
Shan Zhou
WLWeigang Lu
Weigang Lu
In recent years, Generative Artificial Intelligence (GAI) has demonstrated remarkable potential for producing creative content, including artwork, which has sparked discussions regarding its role in education. Although AI-generated art is gaining recognition in the art world, its application in visual art education remains underexplored. To address this gap, we focused on AI-generated images-a specific category within AI-generated art that emphasizes visual representation. We integrated these images into visual art education and conducted an experiment to evaluate their effects. Seventy-eight fifth-grade students were randomly assigned to a treatment group (n = 39) and a control group (n = 39). Initially, both groups received conventional visual art instruction using classical images. Subsequently, the treatment group was introduced to a GAI-assisted teaching method utilizing AI-generated images, while the control group continued with conventional instruction. The results revealed that students in the treatment group exhibited significantly higher levels of classroom engagement compared to their peers in the control group. Moreover, the treatment group reported a strong sense of self-efficacy with the GAI-assisted method. Importantly, there were no significant differences in cognitive load between the two groups. A comparative analysis of the students’ paintings was conducted, focusing on technical skill, adherence to theme, composition and design, creativity and originality, effort, and improvement, which aligns with the increased classroom engagement and self-efficacy, thereby supporting the effectiveness of AI-generated images. This study is one of the pioneering works to propose and validate the use of AI-generated images to address challenges in developing learning materials within the context of visual art education.

14 February 2023 in Computer Vision and Image Understanding
Verified
Journal Article
Closed access
CBCunling Bian
Cunling Bian
WFWei Feng
Wei Feng
FMFanbo Meng
Fanbo Meng
SWSong Wang
Song Wang

16 April 2025 in Scientific Data
Verified
Journal Article
Open access
PDF file
WLWeigang Lu
Weigang Lu
YYYang Yang
Yang Yang
RSRunfei Song
Runfei Song
YCYing Chen
Ying Chen
TWTao Wang
Tao Wang
CBCunling Bian
Cunling Bian
Student group engagement helps share knowledge and build a more complete understanding. Recognition of group engagement in the classroom helps us understand students’ learning state and optimize teaching and study processes. While existing research predominantly focuses on individual engagement recognition through controlled lab-based computer interactions, this paradigm was fundamentally misaligned with authentic classroom dynamics. To address this issue, this research focuses on student group engagement recognition in classroom environments using visual cues. We propose OUC Classroom Group Engagement Dataset (OUC-CGE), the first benchmark dedicated to group engagement analysis in authentic classroom settings using pure visual signals. Several classical models are tested on OUC-CGE, and through technical-pedagogical dual validation strategy, OUC-CGE exhibits good consistency and discriminability with existing datasets. By transcending the individualistic paradigm, this work establishes group engagement as a computable pedagogical construct, offering teachers diagnostic insights into group engagement trajectories while preserving the ecological complexity of authentic classrooms. The OUC-CGE dataset and models are publicly released to catalyze research in socially-embedded educational artificial intelligence.

16 May 2025 in Sensors
Verified
Journal Article
Open access
CBCunling Bian
Cunling Bian
YYYang Yang
Yang Yang
TWTao Wang
Tao Wang
WLWeigang Lu
Weigang Lu
Skeleton representation learning offers substantial advantages for action recognition by encoding intricate motion details and spatial–temporal dependencies among joints. However, fully supervised approaches necessitate large amounts of annotated data, which are often labor-intensive and costly to acquire. In this work, we propose the Spatial–Temporal Heatmap Masked Autoencoder (STH-MAE), a novel self-supervised framework tailored for skeleton-based action recognition. Unlike coordinate-based methods, STH-MAE adopts heatmap volumes as its primary representation, mitigating noise inherent in pose estimation while capitalizing on advances in Vision Transformers. The framework constructs a spatial–temporal heatmap (STH) by aggregating 2D joint heatmaps across both spatial and temporal axes. This STH is partitioned into non-overlapping patches to facilitate local feature learning, with a masking strategy applied to randomly conceal portions of the input. During pre-training, a Vision Transformer-based autoencoder equipped with a lightweight prediction head reconstructs the masked regions, fostering the extraction of robust and transferable skeletal representations. Comprehensive experiments on the NTU RGB+D 60 and NTU RGB+D 120 benchmarks demonstrate the superiority of STH-MAE, achieving state-of-the-art performance under multiple evaluation protocols.

19 January 2026 in IEEE Transactions on Affective Computing
Verified
Journal Article
Closed access
WLWeigang Lu
Weigang Lu
YCYing Chen
Ying Chen
YYYang Yang
Yang Yang
RSRunfei Song
Runfei Song
TWTao Wang
Tao Wang
CBCunling Bian
Cunling Bian
Classroom climate is a socio-emotional environment shaped by teacher–student interactions, directly influencing classroom practices and student development. However, the diverse behaviors, subtle emotional cues, and complex interaction structures present in real classrooms make it difficult to develop intelligent models that can accurately perceive and interpret this environment. Existing methods often rely on single-modality features or generic multimodal models, and thus fail to represent the distinctive characteristics of classroom groups, resulting in shallow or unstable climate recognition. To effectively capture the dynamic, multimodal patterns of real-world classrooms, we introduce the Multimodal Classroom Climate Dataset (MCCD), which includes a wide range of instructional activities, and propose the Multimodal Classroom Climate Recognition Network (MCCRNet). The network integrates visual cues such as skeletal actions with auditory features such as emotional tone to construct cross-modal representations that align with the logic of classroom interactions. Experimental results indicate that the proposed approach exceeds existing methods, laying a preliminary groundwork for advancing the understanding of classroom ecology and instructional decision-making.