arXiv subjects

Jingsheng Li

Publications and source records attributed to Jingsheng Li.

5 recordsLinked to original sources

arXiv2026

CloudMatch: Weak-to-Strong Consistency Learning for Semi-Supervised Cloud Detection

Due to the high cost of annotating accurate pixel-level labels, semi-supervised learning has emerged as a promising approach for cloud detection. In this paper, we propose CloudMatch, a semi-supervised framework that effectively leverages unlabeled remote sensing imagery through view-consistency learning combined with scene-mixing augmentations. An observation behind CloudMatch is that cloud patterns exhibit structural diversity and contextual variability across different scenes and within the same scene category. Our key insight is that enforcing prediction consistency across diversely augmented views, incorporating both inter-scene and intra-scene mixing, enables the model to capture the structural diversity and contextual richness of cloud patterns. Specifically, CloudMatch generates one weakly augmented view along with two complementary strongly augmented views for each unlabeled image: one integrates inter-scene patches to simulate contextual variety, while the other employs intra-scene mixing to preserve semantic coherence. This approach guides pseudolabel generation and enhances generalization. Extensive experiments show that CloudMatch achieves good performance, demonstrating its capability to utilize unlabeled data efficiently and advance semi-supervised cloud detection.

cs.CV↗

arXiv2025

CD-Mamba: Cloud detection with long-range spatial dependency modeling

Remote sensing images are frequently obscured by cloud cover, posing significant challenges to data integrity and reliability. Effective cloud detection requires addressing both short-range spatial redundancies and long-range atmospheric similarities among cloud patches. Convolutional neural networks are effective at capturing local spatial dependencies, while Mamba has strong capabilities in modeling long-range dependencies. To fully leverage both local spatial relations and long-range dependencies, we propose CD-Mamba, a hybrid model that integrates convolution and Mamba's state-space modeling into a unified cloud detection network. CD-Mamba is designed to comprehensively capture pixelwise textural details and long term patchwise dependencies for cloud detection. This design enables CD-Mamba to manage both pixel-wise interactions and extensive patch-wise dependencies simultaneously, improving detection accuracy across diverse spatial scales. Extensive experiments validate the effectiveness of CD-Mamba and demonstrate its superior performance over existing methods.

cs.CV↗

arXiv2025

Uncertainty modeling method for wind and solar power output in building integrated energy systems under continuous anomalous weather

The increasing occurrence of continuous anomalous weather events has intensified the uncertainty in wind and photovoltaic power generation, posing significant challenges to the operation and optimization of building integrated energy systems. Existing studies often neglect the interdependence between successive anomalous weather events and their collective impact on wind and solar power output. Additionally, conventional modeling approaches struggle to accurately capture the nonlinear fluctuations induced by these weather conditions. To address this gap, this study proposes an uncertainty modeling method based on stochastic optimization and scenario generation. The Weibull and Beta distributions characterize the probabilistic properties of wind speed and solar irradiance, respectively, while the Copula function captures the dependence between wind speed and precipitation, enabling the construction of a wind-solar power uncertainty model that incorporates the joint distribution of consecutive anomalous weather events. A Monte Carlo-based scenario generation approach is employed to construct a dataset representing anomalous weather characteristics, followed by a probabilistic distance-based scenario reduction technique to enhance modeling efficiency. Furthermore, the unscented transformation method is introduced to mitigate nonlinear propagation errors in wind and solar power state estimation. Case studies demonstrate that the proposed method effectively characterizes the fluctuation patterns of wind and solar power under continuous anomalous weather conditions while preserving the statistical properties of the original data. These findings provide a reliable basis for improving the operational resilience of building integrated energy systems under extreme weather scenarios.

math.OC↗

arXiv2024

High-Resolution Cloud Detection Network

The complexity of clouds, particularly in terms of texture detail at high resolutions, has not been well explored by most existing cloud detection networks. This paper introduces the High-Resolution Cloud Detection Network (HR-cloud-Net), which utilizes a hierarchical high-resolution integration approach. HR-cloud-Net integrates a high-resolution representation module, layer-wise cascaded feature fusion module, and multi-resolution pyramid pooling module to effectively capture complex cloud features. This architecture preserves detailed cloud texture information while facilitating feature exchange across different resolutions, thereby enhancing overall performance in cloud detection. Additionally, a novel approach is introduced wherein a student view, trained on noisy augmented images, is supervised by a teacher view processing normal images. This setup enables the student to learn from cleaner supervisions provided by the teacher, leading to improved performance. Extensive evaluations on three optical satellite image cloud detection datasets validate the superior performance of HR-cloud-Net compared to existing methods.The source code is available at \url{https://github.com/kunzhan/HR-cloud-Net}.

cs.CV↗

arXiv2011

Supersymmetric Nonlinear Sigma Model in AdS_5

We construct the supersymmetric nonlinear sigma model in a fixed AdS_5 background. We use component fields and find that the complex bosons must be the coordinates of a hyper-Kahler manifold that admits a Killing vector satisfying an inhomogeneous tri-holomorphic condition. We propose boundary conditions that map the on-shell bulk hypermultiplets into off-shell chiral multiplets on 3-branes that foliate the bulk. The supersymmetric AdS_5 isometries reduce to superconformal transformations on the brane fields.

hep-th↗