Understanding flood scenes is essential for effective disaster response. Previous research has primarily focused on computer vision-based approaches for analyzing flood scenes, capitalizing on their ability to rapidly and accurately cover affected regions. However, most existing methods emphasize static image analysis, with limited attention given to dynamic video analysis. Compared to image-based approaches, video analysis in flood scenarios offers significant advantages, including real-time monitoring, flow estimation, object tracking, change detection, and behavior recognition. To address this gap, this study proposes a computer vision-based multi-object tracking (MOT) framework for intelligent flood scene understanding. The proposed method integrates an optical-flow-based module for short-term undetected mask estimation and a deep re-identification (ReID) module to handle long-term occlusions. Experimental results demonstrate that the proposed method achieves state-of-the-art performance across key metrics, with a HOTA of 69.57%, DetA of 67.32%, AssA of 73.21%, and IDF1 of 89.82%. Field tests further confirm its improved accuracy, robustness, and generalization. This study not only addresses key practical challenges but also offers methodological insights, supporting the application of intelligent technologies in disaster response and humanitarian aid.

Photovoltaic double-skin façades (PV-DSFs) can block solar radiation heat, mitigate air heat transfer, facilitate ventilation cooling, and generate electricity, making them a high-performance building envelope suitable for hot southern regions in summer. The thermal performance of DSFs is relatively well understood; however, with the addition of photovoltaic glass panels, the influence of design parameters is altered due to thermoelectric coupling effects. Then, the influence of design parameters on their thermoelectric performance remains unclear, hindering their design optimization. This paper establishes a mathematical model for DSFs with MATLAB (R2023a) to analyze their thermoelectric performance and the impact of design parameters. The results indicate that the daily power generation of PV-DSFs is primarily influenced by the solar radiation on the west-facing vertical surface. The wall exterior surface gains heat via longwave radiation during the day and loses heat at night, while convective heat dissipation occurs throughout the entire day, with radiative heat flux being the dominant mechanism. The power generation of photovoltaic cells is significantly influenced by their coverage ratio, while the impact of other factors can be neglected. The temperature of the wall’s exterior surface is significantly influenced by the heat storage of the outer cladding panel, the solar absorptivity of the exterior surface, and the emissivity of the interior surface. Among these factors, the heat storage of the outer cladding panel primarily affects the attenuation and delay of peak values and temperature fluctuations on the exterior surface. Meanwhile, the solar absorptivity of the exterior surface and the emissivity of the interior surface mainly influence the peak temperature of the wall’s exterior surface, with the effect becoming more pronounced when the interior surface emissivity is lower.

Artificial Intelligence (AI) has emerged as a transformative force in the field of architectural design. This study aims to systematically analyze the influence mechanisms of Artificial Intelligence-based Architectural Design (AIAD) by constructing a comprehensive hybrid model that integrates the Analytic Hierarchy Process (AHP), Decision-Making Trial and Evaluation Laboratory (DEMATEL), Interpretive Structural Modeling (ISM), and Cross-Impact Matrix Multiplication Applied to Classification (MICMAC). Based on the previous quantitative literature review, 6 primary categories and 18 secondary influencing factors were identified. Data were collected from a panel of fifteen experts representing the architecture industry, academia, and computer science. Through weighting analysis, causal mapping, hierarchical structuring, and driving–dependence classification, the study clarifies the complex interrelationships among influencing factors and reveals the underlying drivers that accelerate or constrain AI adoption in architectural design. By quantifying the hierarchical and causal influence of factors, this research provides theoretical findings and practical insights for design firms undergoing digital transformation. The results extend previous meta-analytical studies, offering a decision-support system that bridges academic research and real-world applications, thereby guiding stakeholders toward informed adoption of artificial intelligence for future cultural tourism development and regional spatial innovation.

Integrating artificial intelligence (AI) into residential architectural design faces challenges due to fragmented workflows and the lack of localized datasets. This study proposes the CorbuAI framework, hypothesizing that a multimodal AI system integrating Pix2pix-GAN and Stable Diffusion (SD) can streamline the transition from floor plan generation to elevation and interior design within a specific regional context. We developed a custom dataset featuring 2335 manually refined Chinese residential floor plans and 1570 elevation images. The methodology employs a specialized U-Net V2.0 generator for functional layout synthesis and an SD-based model for stylistic transfer and elevation rendering. Evaluation was conducted through both subjective professional scoring and objective metrics, including the Perceptual Hash Algorithm (pHash). Results demonstrate that CorbuAI achieves high accuracy in spatial allocation (scoring 0.88/1.0) and high structural consistency in elevation generation (mean pHash similarity of 0.82). The framework significantly reduces design iteration time while maintaining professional aesthetic standards. This research provides a scalable AI-driven methodology for automated residential design, bridging the gap between schematic layouts and visual representation in the Chinese architectural context.

Despite significant global commitment to smart buildings and Digital Twin technologies within architecture research and practice, existing systems face fundamental challenges: they widely suffer from data silos, hindering comprehensive data integration; they are constrained by cognitive limitations, preventing deep learning from predicting complex behaviors and spatial intentions; their design goals are fundamentally device-centric, rather than human-centric; and they operate in a state of environmental isolation, lacking dynamic coordination with the external environment. To address these bottlenecks, this paper proposes the BLUE Building Paradigm, a novel and pioneering framework for next-generation spatial intelligence. BLUE represents four core pillars: Big-data (B), Learning (L), User (U), and Environment (E). The core contribution of the BLUE Building is the construction of a Spatio-Temporal Cognitive Operating System, which, through a unified Spatial Semantic Graph and Adaptive Reinforcement Learning, achieves a deep understanding of spatial states and user intentions, and forms a dynamic, continuously optimized closed-loop synergy with the external environment. This marks an epoch-making transition in spatial intelligence from passive automation to proactive cognition and continuous self-adaptation. This paper details the operational mechanism of the BLUE Paradigm, its key technical implementations, and cross-scale interaction strategies. Furthermore, it introduces the BLUE Building Rating and Evaluation Mechanism, including its potential for expansion, translating the paradigm's cognitive capabilities into quantifiable industry standards to drive adoption. The BLUE Building Paradigm not only sets a new benchmark for building energy efficiency and occupant well-being but also lays a solid theoretical foundation for the resilient, sustainable, and integrated development of future urban systems and the human experience.