Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kaixin Chen has a total of 23 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • From Sight to Insight: Enhancing Confusable Structure Segmentation via Vision-Language Mutual Prompting

    by Yixiang Huang, Yihao Zuo, Mengqiu Xu, Kaixin Chen, Ming Wu, Chuang Zhang, Zhanyu Ma, Jun Guo

    Confusable structure segmentation (CSS) is a type of semantic segmentation applied in remote sensing sea fog detection, medical image segmentation, camouflaged object detection, etc. Structural similarity and visual ambiguity are two critical issues in CSS that pose difficulties in distinguishing foreground objects from the background. Current methods focus primarily on enhancing visual representations and do not often incorporate multimodal information, which leads to performance bottlenecks. Inspired by recent achievements in vision-language models, we propose Vision-Language Mutual Prompting (VLMP), a novel and unified language-guided framework that leverages text prompts to enhance CSS. Specifically, VLMP consists of vision-to-language prompting and language-to-vision prompting, which bidirectionally model the interactions between visual and linguistic features, thereby facilitating cross-modal complementary information flow. To prevent the predominance of one modality over another, we design a feature integration modulator that modulates and balances feature weights for adaptive multimodal fusion. Our framework is designed to be modular and flexible, allowing for integration with any backbone, including CNNs and transformers. We evaluate VLMP with three diverse datasets: SFDD-H8, QaTa-COV19, and CAMO-COD10K. Extensive experiments demonstrate the effectiveness and superiority of the proposed framework over those of state-of-the-art methods across these datasets. This shift from basic sight to deeper insight in CSS through vision-language integration represents a significant advancement in the field.

  • Open Access

    SirP: Semi-Recurrent Paradigm-Based Prediction Network for Satellite Image Sequences

    by Kaixin Chen, Mengqiu Xu, Weiqing Li, Fanbin Mo, Yixiang Huang, Ming Wu, Chuang Zhang

    Satellite image sequence prediction, which supports critical applications in disaster prevention and environmental monitoring, remains challenging due to the detail-rich and long-range chaotic nature of satellite data. Existing data-driven paradigms-whether recurrent or non-recurrent-struggle to effectively capture both short- and long-term spatiotemporal dynamics. To address this, we design a novel semi-recurrent paradigm that combines parallel input encoding with a recurrent, progressive output mechanism, enabling robust reasoning and prediction for both short and long sequences. In addition, we propose SirP, a new prediction network featuring three key innovations specifically designed for the semi-recurrent paradigm and the characteristics of satellite imagery: a triple-view attention module, an accumulated-and-current feature-based progressive predictor, and a nearest-frame skip connection. Comprehensive experiments on five datasets constructed from Himawari-8, FY- 4B, and GOES-16 satellite observations, covering the coastal East Asia and North American continental regions, show that SirP consistently outperforms state-of-the-art methods from both recurrent and non-recurrent paradigms across pixel-level and event-level metrics, with greater gains as the prediction horizon extends. Detailed ablation studies confirm the effectiveness of each core component. The code will be made publicly available.


  • Class-Customized Domain Adaptation: Unlock Each Customer-Specific Class with Single Annotation

    by Kaixin Chen, Huiying Chang, Mengqiu Xu, Ruoyi Du, Ming Wu, Zhanyu Ma, Chuang Zhang

    Model customization mitigates the issues of inadequate performance, resource wastage, and privacy risks associated with using general-purpose models in specialized domains and well-defined tasks. However, achieving customization at a low annotation cost still poses a challenge. Existing domain adaptation research has addressed cases where all customized classes are present in the labeled database, yet scenarios involving customer-specific classes are still unresolved. Therefore, this paper proposes a novel Class-Customized Domain Adaptation (CCDA) method, addressing the latter scenario with just one additional annotation for each customer-specific class. CCDA adopts the classic adaptation training framework and comprises two innovative techniques. Firstly, to ensure the shared class knowledge from the database and the private class knowledge from additional annotations are transferred and propagated to the correct regions within the target domain, we design the partial-feature alignment strategy, based on the mechanical properties of feature alignment. Second, we propose soft-balanced sampling to tackle the long-tail distribution problem in labeled data, preventing the model from overfitting to the labeled samples of customer-specific classes. The effectiveness of CCDA has been validated across 48 tasks simulated on domain adaptation benchmarks and two real-world customization scenarios, consistently showing excellent performance. Additionally, extensive analytical experiments illustrate the contributions of two innovative techniques. The code is available at https://github.com/CHEN-kx/ClassCustomizedDA.


  • Open Access

    MFogHub: Bridging Multi-Regional and Multi-Satellite Data for Global Marine Fog Detection and Forecasting

    by Mengqiu Xu, Kaixin Chen, Heng Guo, Yixiang Huang, Ming Wu, Zhenwei Shi, Chuang Zhang, Jun Guo

    Deep learning approaches for marine fog detection and forecasting have outperformed traditional methods, demonstrating significant scientific and practical importance. However, the limited availability of open-source datasets remains a major challenge. Existing datasets, often focused on a single region or satellite, restrict the ability to evaluate model performance across diverse conditions and hinder the exploration of intrinsic marine fog characteristics. To address these limitations, we introduce \textbf{MFogHub}, the first multi-regional and multi-satellite dataset to integrate annotated marine fog observations from 15 coastal fog-prone regions and six geostationary satellites, comprising over 68,000 high-resolution samples. By encompassing diverse regions and satellite perspectives, MFogHub facilitates rigorous evaluation of both detection and forecasting methods under varying conditions. Extensive experiments with 16 baseline models demonstrate that MFogHub can reveal generalization fluctuations due to regional and satellite discrepancy, while also serving as a valuable resource for the development of targeted and scalable fog prediction techniques. Through MFogHub, we aim to advance both the practical monitoring and scientific understanding of marine fog dynamics on a global scale. The dataset and code are at \href{https://github.com/kaka0910/MFogHub}{https://github.com/kaka0910/MFogHub}.

  • Open Access

    DRAF-Net: Dual-Branch Residual-Guided Multi-View Attention Fusion Network for Station-Level Numerical Weather Prediction Correction

    by Kaixin Chen, Jiaxin Chen, Mengqiu Xu, Ming Wu, Chuang Zhang

    Accurate station-level numerical weather predictions are critical for disaster prevention and mitigation, with error correction playing an essential role. However, existing correction models struggle to effectively handle the high-dimensional features and complex dependencies inherent in meteorological data. To address these challenges, this paper proposes the dual-branch residual-guided multi-view attention fusion network (DRAF-Net), a novel deep learning-based correction model. DRAF-Net introduces two key innovations: (1) a dual-branch residual structure that enhances the spatial sensitivity of deep high-dimensional features and improves output stability by connecting raw data and shallow features to deep features, respectively; and (2) a multi-view attention fusion mechanism that models spatiotemporal influences, temporal dynamics, and spatial associations, significantly improving the representation of complex dependencies. The effectiveness of DRAF-Net was validated on two real-world datasets comprising observations and predictions from Chinese meteorological stations. It achieved an average RMSE reduction of 83.44% and an average MAE reduction of 84.21% across all eight variables, significantly outperforming other methods. Moreover, extensive studies confirmed the critical contributions of each key component, while visualization results highlighted the model’s ability to eliminate anomalous values and improve prediction consistency. The code will be made publicly available to support future research and development.


  • Open Access

    Ariadne's Thread:Using Text Prompts to Improve Segmentation of Infected Areas from Chest X-ray images

    by Yi Zhong, Mengqiu Xu, Kongming Liang, Kaixin Chen, Ming Wu

    Segmentation of the infected areas of the lung is essential for quantifying the severity of lung disease like pulmonary infections. Existing medical image segmentation methods are almost uni-modal methods based on image. However, these image-only methods tend to produce inaccurate results unless trained with large amounts of annotated data. To overcome this challenge, we propose a language-driven segmentation method that uses text prompt to improve to the segmentation result. Experiments on the QaTa-COV19 dataset indicate that our method improves the Dice score by 6.09% at least compared to the uni-modal methods. Besides, our extended study reveals the flexibility of multi-modal methods in terms of the information granularity of text and demonstrates that multi-modal methods have a significant advantage over image-only methods in terms of the size of training data required.

  • Adaptive Patch Exiting for Scalable Single Image Super-Resolution

    by Shizun Wang, Jiaming Liu, Kaixin Chen, Xiaoqi Li, Ming Lu, Yandong Guo


  • Open Access

    The Eyes of the Gods: A Survey of Unsupervised Domain Adaptation Methods Based on Remote Sensing Data

    by Mengqiu Xu, Ming Wu, Kaixin Chen, Chuang Zhang, Jun Guo

    With the rapid development of the remote sensing monitoring and computer vision technology, the deep learning method has made a great progress to achieve applications such as earth observation, climate change and even space exploration. However, the model trained on existing data cannot be directly used to handle the new remote sensing data, and labeling the new data is also time-consuming and labor-intensive. Unsupervised Domain Adaptation (UDA) is one of the solutions to the aforementioned problems of labeled data defined as the source domain and unlabeled data as the target domain, i.e., its essential purpose is to obtain a well-trained model and tackle the problem of data distribution discrepancy defined as the domain shift between the source and target domain. There are a lot of reviews that have elaborated on UDA methods based on natural data, but few of these studies take into consideration thorough remote sensing applications and contributions. Thus, in this paper, in order to explore the further progress and development of UDA methods in remote sensing, based on the analysis of the causes of domain shift, a comprehensive review is provided with a fine-grained taxonomy of UDA methods applied for remote sensing data, which includes Generative training, Adversarial training, Self-training and Hybrid training methods, to better assist scholars in understanding remote sensing data and further advance the development of methods. Moreover, remote sensing applications are introduced by a thorough dataset analysis. Meanwhile, we sort out definitions and methodology introductions of partial, open-set and multi-domain UDA, which are more pertinent to real-world remote sensing applications. We can draw the conclusion that UDA methods in the field of remote sensing data are carried out later than those applied in natural images, and due to the domain gap caused by appearance differences, most of methods focus on how to use generative training (GT) methods to improve the model’s performance. Finally, we describe the potential deficiencies and further in-depth insights of UDA in the field of remote sensing.