arXiv subjects

Kuan-Wei Chen

Publications and source records attributed to Kuan-Wei Chen.

2 recordsLinked to original sources

arXiv2026

Synchronization and Hopf Bifurcation in Stuart--Landau Networks

The Kuramoto model has shaped our understanding of synchronization in complex systems, yet its phase-only formulation neglects amplitude dynamics that are intrinsic to many oscillatory networks. In this work, we revisit Kuramoto-type synchronization through networks of Stuart-Landau oscillators, which arise as the universal normal form near a Hopf bifurcation. For identical natural frequencies, we analyze synchronization in two complementary regimes. Away from criticality, we establish exponential complete synchronization on arbitrary finite connected undirected networks under explicit sufficient conditions on the parameters and initial data that prevent amplitude death. For ring networks, we identify an exact branch of synchronous periodic solutions arising from a supercritical Hopf bifurcation and use block-circulant Fourier analysis to determine the critical parameter values and multiplicities of the non-synchronous modes. For $N=7$ and $s=2$, a center-manifold reduction yields cubic amplitude equations for paired critical modes, identifying exact single-mode rotating-wave solutions and a standing-wave pattern at cubic order. Numerical simulations compare the dynamics restricted to the rotating-wave invariant subspaces with direct simulations of the full network.

math.DS↗

arXiv2023

Seg2Reg: Differentiable 2D Segmentation to 1D Regression Rendering for 360 Room Layout Reconstruction

State-of-the-art single-view 360-degree room layout reconstruction methods formulate the problem as a high-level 1D (per-column) regression task. On the other hand, traditional low-level 2D layout segmentation is simpler to learn and can represent occluded regions, but it requires complex post-processing for the targeting layout polygon and sacrifices accuracy. We present Seg2Reg to render 1D layout depth regression from the 2D segmentation map in a differentiable and occlusion-aware way, marrying the merits of both sides. Specifically, our model predicts floor-plan density for the input equirectangular 360-degree image. Formulating the 2D layout representation as a density field enables us to employ `flattened' volume rendering to form 1D layout depth regression. In addition, we propose a novel 3D warping augmentation on layout to improve generalization. Finally, we re-implement recent room layout reconstruction methods into our codebase for benchmarking and explore modern backbones and training techniques to serve as the strong baseline. Our model significantly outperforms previous arts. The code will be made available upon publication.

cs.CV↗