Source author record

Xiaolong Yin

Xiaolong Yin appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Computer Vision cond-mat.soft physics.flu-dyn

Catalog footprint

What is connected

3works

3topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

Img2CADSeq: Image-to-CAD Generation via Sequence-Based Diffusion

Boundary Representation (BRep) is the standard format for Computer-Aided Design (CAD), yet reconstructing high-quality BReps from single-view images remains challenging due to the complexity of topological constraints and operation sequences. We present Img2CADSeq, a multi-stage pipeline that overcomes these limitations by encoding CAD sequences into a three-level hierarchical codebook. Guided by an importance prioritization, this strategy values profiles over details, compressing long sequences into a stable discrete latent space. To bridge the modality gap, we leverage a coarse-to-fine point cloud intermediate, aligning 2D visual features with 3D CAD sequences via contrastive learning to condition a VQ-Diffusion model. Supported by newly introduced CAD-220K and PrintCAD datasets, our approach ensures robust industrial domain adaptation. Extensive experiments demonstrate that Img2CADSeq significantly outperforms state-of-the-art methods, producing standard STEP files that can be directly used in commercial CAD software.

preprint2016arXiv

Transport coefficients of solid particles immersed in a viscous gas

Transport properties of a suspension of solid particles in a viscous gas are studied. The dissipation in such systems arises from two sources: inelasticity in particle collisions and viscous dissipation due to the effect of the gas phase on the particles. Here, we consider a simplified case in which the mean relative velocity between the gas and solid phases is taken to be zero, such that "thermal drag" is the only remaining gas-solid interaction. Unlike the previous more general treatment of the drag force [Garzó \emph{et al.}, J. Fluid Mech. \textbf{712}, 129 (2012)], here we take into account contributions to the (scaled) transport coefficients $η^*$ (shear viscosity), $κ^*$ (thermal conductivity) and $μ^*$ (Dufour-like coefficient) coming from the temperature-dependence of the (dimensionless) friction coefficient $γ^*$ characterizing the amplitude of the drag force. At moderate densities, the thermal drag model (which is based on the Enskog kinetic equation) is solved by means of the Chapman-Enskog method and the Navier-Stokes transport coefficients are determined in terms of the coefficient of restitution, the solid volume fraction and the friction coefficient. The results indicate that the effect of the gas phase on $η^*$ and $μ^*$ is non-negligible (especially in the case of relatively dilute systems) while the form of $κ^*$ is the same as the one obtained in the dry granular limit. Finally, as an application of these results, a linear stability analysis of the hydrodynamic equations is carried out to analyze the conditions for stability of the homogeneous cooling state. A comparison with direct numerical simulations shows a good agreement for conditions of practical interest.

preprint2012arXiv

Water-oil drainage dynamics in oil-wet random microfluidic porous media analogs

Displacement experiments carried out in microfluidic porous media analogs show that reduced surface tension leads to a more stable displacement, opposite to the process in Hele-Shaw cells where surface tension stabilizes the displacement of a more viscous fluid by a less viscous fluid. In addition, geometry of porous media is observed to play an important role. Three random microfluidic porous media analogs were made to study water-oil drainage dynamics, featuring a pattern of randomly connected channels with a uniform width, a pattern with Gaussian channel width distribution, and a pattern with large isolated pores. The microfluidic chips fabricated using Polydimenthylsiloxane with glass covers have the internal surface treated by Trichlorosilane to achieve a uniform oil-wet condition. The aqueous phase displaces the oil phase, with a viscosity ratio of about 1:40 and a density ratio of 1:0.85. Videos 1-3 show water flooding processes. It is observed that both channel size distribution (Video 2) and heterogeneity in pore size (Video 3) lead to stronger fingers and reduced displacement efficiency. Video 4 shows that meniscus in small channels retreat as water front moves into a nearby large cavity due to the disparity in the capillary force and contact angle hysteresis. Videos 5 and 6, both taken at 100X magnification in Chip 2, show the stabilizing effect of reduced interfacial tension.