Chuangxin Cai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Chuangxin Cai has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Xianxuan Lin
Nanjing University of Information Science and Technology, Nanjing, China
4 shared publications - Zhigeng Pan
Nanjing University of Information Science and Technology
4 shared publications - Aditi Bhattarai
Nanjing University of Information Science and Technology, Nanjing, China
3 shared publications - Jing Zhang
School of Artificial Intelligence, Nanjing University of Information Science and Technology, Nanjing, China
2 shared publications - Kao Zhang
The School of Artificial Intelligence/School of Future Technology, Nanjing University of Information Science and Technology, Nanjing 210044, China
2 shared publications - Bailin Yang
Zhejiang Gongshang University
1 shared publication
Latest Publications
- Open Access
- 12 January 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (1) , 1-24
MambaWDC: Efficient Weather Data Compression via Selective State Space Model
The storage and transmission of large-scale meteorological data have become a significant bottleneck hindering the in-depth development of meteorological research. This study proposes a novel meteorological data compression framework characterized by innovative thinking. It embeds the Mamba structure during the feature extraction stage and integrates a spatio-temporal channel attention mechanism to construct an entropy coding mechanism that facilitates the sharing of high-dimensional meteorological data channel information. The framework achieves a compression rate of 300x and, more importantly, preserves the essential spatio-temporal features of meteorological data. Validation on the ERA5 dataset demonstrates that the proposed method surpasses existing techniques in terms of both data compression efficiency and reconstruction accuracy. The accuracy of meteorological analyses based on compressed data is comparable to that of the original data, offering a highly practical solution to the challenges of storing meteorological big data. Our source code is publicly available online at https://github.com/cike0cop/MambaComp .
- 12 January 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
- Vol. 22 (1) , 1-24
Open AccessMambaWDC: Efficient Weather Data Compression via Selective State Space Model
The storage and transmission of large-scale meteorological data have become a significant bottleneck hindering the in-depth development of meteorological research. This study proposes a novel meteorological data compression framework characterized by innovative thinking. It embeds the Mamba structure during the feature extraction stage and integrates a spatio-temporal channel attention mechanism to construct an entropy coding mechanism that facilitates the sharing of high-dimensional meteorological data channel information. The framework achieves a compression rate of 300x and, more importantly, preserves the essential spatio-temporal features of meteorological data. Validation on the ERA5 dataset demonstrates that the proposed method surpasses existing techniques in terms of both data compression efficiency and reconstruction accuracy. The accuracy of meteorological analyses based on compressed data is comparable to that of the original data, offering a highly practical solution to the challenges of storing meteorological big data. Our source code is publicly available online at https://github.com/cike0cop/MambaComp .
- Open Access
- 13 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-6
ALTSA: Adaptive Latent-space Teacher - Student Architecture for Emotion Recognition in Conversation
- 13 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessALTSA: Adaptive Latent-space Teacher - Student Architecture for Emotion Recognition in Conversation
- 26 May 2025
- Published by Elsevier in Neurocomputing
- Vol. 647
Prompt-based hybrid supervised contrastive learning for emotion recognition in conversation
- 26 May 2025
- Published by Elsevier in Neurocomputing
- Vol. 647
Prompt-based hybrid supervised contrastive learning for emotion recognition in conversation
- Open Access
- 1 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Efficient Classroom Behavior Detection through End-to End Multiscale Feature Fusion
Classroom behavior detection is essential for smart education. Enhance teaching effectiveness, promote personalized learning, and support holistic student development. Existing methods that rely on single-feature input are susceptible to noise interference, which can lead to misjudgments in classroom behavior detection. This paper proposes an efficient classroom behavior detection through end-to-end multiscale feature fusion. Initially, the original YOLOv7 input module was replaced with a multi-feature information input module to enhance spatial feature utilization efficiency. Then, a feature weight attention mechanism was incorporated to allow the main network to weigh feature information on various scales comprehensively. This approach has improved the accuracy of the algorithm’s identification and reduced the number of network parameters. Finally, including a focal loss function suppressed background interference, improving recognition accuracy. Experimental results demonstrate that the improved algorithm attains 94.1% accuracy in the RizeHand dataset and effectively recognizes student behaviors in real-world scenarios.
- 1 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessEfficient Classroom Behavior Detection through End-to End Multiscale Feature Fusion
Classroom behavior detection is essential for smart education. Enhance teaching effectiveness, promote personalized learning, and support holistic student development. Existing methods that rely on single-feature input are susceptible to noise interference, which can lead to misjudgments in classroom behavior detection. This paper proposes an efficient classroom behavior detection through end-to-end multiscale feature fusion. Initially, the original YOLOv7 input module was replaced with a multi-feature information input module to enhance spatial feature utilization efficiency. Then, a feature weight attention mechanism was incorporated to allow the main network to weigh feature information on various scales comprehensively. This approach has improved the accuracy of the algorithm’s identification and reduced the number of network parameters. Finally, including a focal loss function suppressed background interference, improving recognition accuracy. Experimental results demonstrate that the improved algorithm attains 94.1% accuracy in the RizeHand dataset and effectively recognizes student behaviors in real-world scenarios.
- Open Access
- 1 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-9
FP-KDNet: Facial Perception and Knowledge Distillation Network for Emotion Recogniton in Coversation
Emotion recognition in conversation (ERC) is anchored in the burgeoning field of artificial intelligence, aiming to equip machines with the ability to discern and respond to human emotions in nuanced ways. However, recent studies have primarily focused on textual modalities, often neglecting the significant potential of non-verbal cues found in audio and video, which are critical for accurately capturing emotions. Furthermore, when researchers integrate these non-verbal cues into multimodal emotion recognition systems, they encounter challenges related to data heterogeneity across different modalities. This paper introduces the Facial Perception and Knowledge Distillation Network (FP-KDNet) to address these challenges. Specifically, a novel Facial Perceptual Attention (FPA) module was designed to capture non-verbal cues from videos, significantly enhancing the model’s ability to process visual information. Additionally, a knowledge distillation (KD) strategy was proposed to improve emotion representation in the non-verbal modality by leveraging data from the text modality, facilitating effective cross-modal information exchange. A multi-head attention mechanism further optimizes the integration of features across modalities, dynamically adjusting attention allocation to enhance conversational emotion recognition. The experimental results demonstrated that FP-KDNet achieves excellent performance on the MELD and IEMOCAP datasets, and ablation studies confirm the effectiveness of the multimodal fusion approach.
- 1 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-9
Open AccessFP-KDNet: Facial Perception and Knowledge Distillation Network for Emotion Recogniton in Coversation
Emotion recognition in conversation (ERC) is anchored in the burgeoning field of artificial intelligence, aiming to equip machines with the ability to discern and respond to human emotions in nuanced ways. However, recent studies have primarily focused on textual modalities, often neglecting the significant potential of non-verbal cues found in audio and video, which are critical for accurately capturing emotions. Furthermore, when researchers integrate these non-verbal cues into multimodal emotion recognition systems, they encounter challenges related to data heterogeneity across different modalities. This paper introduces the Facial Perception and Knowledge Distillation Network (FP-KDNet) to address these challenges. Specifically, a novel Facial Perceptual Attention (FPA) module was designed to capture non-verbal cues from videos, significantly enhancing the model’s ability to process visual information. Additionally, a knowledge distillation (KD) strategy was proposed to improve emotion representation in the non-verbal modality by leveraging data from the text modality, facilitating effective cross-modal information exchange. A multi-head attention mechanism further optimizes the integration of features across modalities, dynamically adjusting attention allocation to enhance conversational emotion recognition. The experimental results demonstrated that FP-KDNet achieves excellent performance on the MELD and IEMOCAP datasets, and ablation studies confirm the effectiveness of the multimodal fusion approach.