Beijing Union University

Generative Artificial Intelligence (Generative AI) is transforming how travelers plan trips, yet research has not fully examined dynamic interactions between travelers and generative AI or the co-creation of authenticity through these interactions. This exploratory qualitative study employs a retrospective think-aloud method to explore how travelers engage with generative AI. Grounded in the lenses of host-guest relations and authenticity, this study conceptualizes travel planning as a co-performance between the digital guest (the traveler) and the digital host (generative AI). Findings identify six AI roles (mentor, pathfinder, partner, acquaintance, butler, and implementer) shaped by power dynamics and encounter depth, and four authenticity pathways (co-creation, co-recovery, co-reduction, and co-destruction) across five pre-travel experience stages: triggering inspiration, shaping vision, gathering details, evaluating information, and making decisions. The study advances a relational and performative framework for understanding human-AI collaboration in tourism and offers practical insights for enhancing AI-assisted destination marketing and traveler experiences.

The focus settings and physical characteristics of lenses significantly affect distortion correction, particularly under blurred imaging conditions, thereby limiting the measurement accuracy of existing stereo vision systems. Moreover, traditional calibration processes for stereo vision systems are complex due to the need for precise focus distance adjustments and careful calibration board selection. In this paper, we propose a camera parametric model based on weighted radial constraints that accounts for both radial and non-central lens distortion. This model aims to improve calibration accuracy while simplifying the calibration workflow. Additionally, we introduce a stereo vision correction method that utilizes speckle patterns as calibration images and eliminates the requirement for precise camera focusing. This approach is more user-friendly than conventional planar chessboard techniques, as it avoids frequent lens adjustments for optimal focus. Experimental results demonstrate improvements in accuracy of 17.2% and 15.6% over Zhang’s and Tsai’s calibration methods, respectively.

The system architecture adopts the B/S (Browser/Server) structure. The client side does not require installation of any specialized software; a web browser is sufficient to access and use the monitoring system, enabling energy consumption monitoring anytime and anywhere. It offers good scalability, as future functional expansions and upgrades to the software only need to be implemented on the server side. Anyone can browse and use this system, making energy-saving promotion more extensive and effective.

Accurate and robust visual localization under changing environments remains a fundamental challenge in autonomous driving and mobile robotics. Traditional handcrafted features often degrade under long-term illumination and viewpoint variations, while recent CNN-based methods, although more robust, typically rely on coarse semantic cues and remain vulnerable to dynamic objects. In this paper, we propose a fine-grained semantics-guided feature extraction framework that adaptively selects stable keypoints while suppressing dynamic disturbances. A fine-grained semantic refinement module subdivides coarse semantic categories into stability-homogeneous sub-classes, and a dual-attention mechanism enhances local repeatability and semantic consistency. By integrating physical priors with self-supervised clustering, the proposed framework learns discriminative and reliable feature representations. Extensive experiments on the Aachen and RobotCar-Seasons benchmarks demonstrate that the proposed approach achieves state-of-the-art accuracy and robustness while maintaining real-time efficiency, effectively bridging coarse semantic guidance with fine-grained stability estimation. Quantitatively, our method achieves strong localization performance on Aachen (up to 88.1% at night under the (0.2°,0.25 m) threshold) and on RobotCar-Seasons (up to 57.2%/28.4% under the same threshold for day/night), demonstrating improved robustness to seasonal and illumination changes.

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized primarily by deterioration in memory, cognition, and learning ability. Its etiology is complex and influenced by multiple factors, including genetics and environment. With advancing research into mitochondrial function and mechanisms, impaired mitophagy has been proposed as a significant mechanism contributing to AD. The ApoE ε4 allele, a high-risk genetic factor for AD, may play a key role in disease pathogenesis by inducing mitophagy dysfunction and apoptosis. From the perspective of APOE gene polymorphisms, this study investigates abnormal changes in mitochondrial function and autophagy in humanized APOE4 mice primary astrocytes under oxidative stress, as well as the regulatory effect of curcumin (Cur) on mitophagy and oxidative stress-induced apoptosis, thereby exploring its potential to ameliorate AD through targeting mitophagy. Mitochondrial function analysis revealed that APOE4 expression reduced the antioxidant capacity and respiratory function of primary astrocytes, leading to mitochondrial membrane damage, intracellular reactive oxygen species (ROS) accumulation, and decreased ATP production. Curcumin effectively protected mitochondrial integrity, reduced the number of damaged mitochondria, improved overall mitochondrial function, and helped maintain mitochondrial homeostasis involving in PINK1/Parkin pathway. Regarding autophagy and apoptosis, curcumin was shown to restore autophagic flux, mitigate autophagy disruption caused by oxidative stress, and reverse early-stage apoptosis.