Juntao Yang

DA-NBV: A Direction-Aware Next-Best-View Planner for Efficient 3D Reconstruction of Ships at Sea

Jiaming Chen, Juntao Yang, Zhentao Zou, Qi Ming et autres

Accurate 3D reconstruction of ships at sea is important for maritime supervision, damage assessment, and autonomous maritime operations. Although 3D reconstruction has advanced considerably, high-quality data acquisition still largely relies on manually designed trajectories or skilled operators, resulting in high costs and …

A Coarse-to-Fine Indoor Point Cloud Registration Method Guided by Prior Correspondences

Mengchong Sun, Juntao Yang, Yutao Zhang, Sa Li et autres

Abstract. Superpoint matching is a critical step in coarse-to-fine point cloud registration, and its performance directly affects the accuracy of subsequent point matching and pose estimation. However, most existing methods establish correspondences mainly relying on feature similarity, without explicit modeling of spatial …

cn (code pays fourni par la source)

Text-Guided Semantic Segmentation Method for Indoor 3D Point Clouds

Jinyu Tan, Juntao Yang, Yutao Zhang, Sa Li et autres

Abstract. Point cloud semantic segmentation of indoor environments is a fundamental task in 3D scene understanding. However, existing methods mainly rely on geometric structures and color information, which are prone to error results in scenarios involving occlusion, sparse sampling, and geometrically similar …

cn (code pays fourni par la source)

High-Throughput Estimation of Aboveground Peanut Biomass with Optimized Spectral–Textural Features from Unmanned Aerial Vehicle Multispectral Imagery

Liya Hu, Bo Bai, Juntao Yang, Mingxuan Song et autres

cn (code pays fourni par la source)

NP4Buoyancy: An end-to-end differentiable neural-physics framework for buoyancy-driven flows

Yu Jiao, H L Zhang, Tingkai Xue, Boyang Chen et autres

sg, gb (code pays fourni par la source)

SyneFloor: Synergizing zero-shot geometric priors and active learning for weakly supervised floorplan reconstruction

Xuzhe Wang, Chen Geng, Hao Yu, Zhiyan Ping et autres

cn (code pays fourni par la source)

NeuralFVM: Neural-physics-based Finite Volume Method for Turbulent Flows Using the $k$-$ω$ Model

Tingkai Xue, Yu Jiao, Te Ba, Jingliang Wang et autres

In this work, we develop a neural-physics solver based on finite volume method (FVM), namely NeuralFVM, for turbulent flows by implementing the standard $k$-$ω$ model designed for efficient Graphics Processing Unit (GPU) execution. The governing equations for fluid flow and heat transfer …

sg, gb (code pays fourni par la source)

Geometric phase wavefront manipulation for arbitrary polarization states based on metasurfaces

Dong Hu, Zhiquan Chen, Weiwei Huang, Juntao Yang et autres

超构表面可以在亚波长尺寸内任意调控光子的多个物理维度，为轻量化和小型化光子器件提供了变革性解决方案。在超构表面诸多相位调控机制中，几何相位备受科研人员关注，这归因于其直观且相对容易实现，此外几何相位结构固定的尺寸会降低微纳加工的难度。然而，几何相位的获取伴随着圆偏振态的正交转化，且这一过程局限于光源的圆偏振性。为克服这一局限性，本文提出了一种一般化的波前调控方法，可以在任意的偏振态下实现几何相位调制。这一方法的核心理念是利用偏振调制波片将任意偏振光转换为圆偏振光，并将几何相位赋予给圆偏振光，由此构建起几何相位阵列堆叠在偏振调制阵列上的自由站立式双层超构表面。作为原理的验证，我们在庞加莱球上任意选取了六个偏振态（线偏振、圆偏振和椭圆偏振）并设计了对应的超构表面展示了焦心偏移、轴上变焦和聚焦涡旋等波前调控功能。此外，我们进一步将几何相位与传输相位结合，展示了椭圆偏振复用的全息成像。文章提出的波前调控方法将几何相位拓展到任意的偏振态，为任意偏振态下的波前调控提供了全新的思路。

CondensNet: Self-adaptive physical constraints for stable long-term hybrid climate simulations

Xin Wang, Gianmarco Mengaldo, Jianda Chen, Juntao Yang et autres

Accurate and efficient climate simulations are crucial for understanding Earth’s evolving climate. However, current General Circulation Models (GCMs) face challenges in capturing unresolved physical processes, such as cloud and convection. A common solution is to adopt Cloud-Resolving Models (CRMs), which provide more …

sg, cn, gb, us (code pays fourni par la source)

Study on the time-frequency characteristics of blasting seismic waves and peak particle velocity prediction model in deep mining stope

Yanqi Song, Zhixin Shao, Juntao Yang, Junjie Zheng et autres

To study the signal characteristics and vibration velocity attenuation mechanism of seismic wave in blasting roadway of deep stope, this study systematically analyzes the time-domain, frequency-domain, and instantaneous energy characteristics of seismic waves based on field monitoring data. The results indicate that …

cn (code pays fourni par la source)

Mineral-specific meso-mechanical degradation and microcrack evolution in granite induced by cyclic heating and LN2-cooling

Zhibin Hao, Yanqi Song, Hengyi He, Meilu Yu et autres

LN 2 -assisted cyclic fracturing has been demonstrated to be an effective method for enhancing fracture networks in deep hot dry rock (HDR) geothermal reservoirs. This study investigates the influence of mineral phases on the mesoscopic mechanical response and microcrack evolution in …

cn (code pays fourni par la source)

Macroscopic and Microscopic Mechanical Responses and Correlation of Granite Subjected to Heating and LN2-cooling Cycles

Zhibin Hao, Yanqi Song, Zhonghu Wu, Meilu Yu et autres

cn (code pays fourni par la source)