[Submitted on 21 May 2026 (v1), last revised 22 Jul 2026 (this version, v3)]

View PDF HTML (experimental)

Abstract:Novel view synthesis from sparse-view inputs poses a significant challenge in 3D computer vision, particularly for achieving high-quality scene reconstructions with limited viewpoints. We introduce TWINGS, a framework that enhances 3D Gaussian Splatting (3DGS) by directly addressing point sparsity. We employ Thin Plate Splines (TPS), a smooth non-rigid deformation model that minimizes bending energy to estimate a globally coherent warp from control-point correspondences, to align backprojected points from estimated depth with triangulated 3D control points, yielding calibrated backprojected points. By sampling these calibrated points near the control points, TWINGS provides a fast and geometrically accurate initialization for 3DGS, ultimately improving structural detail preservation and color fidelity in reconstructed scenes. Extensive experiments on DTU, LLFF, and Mip-NeRF360 demonstrate that TWINGS consistently outperforms existing methods, delivering detailed and accurate reconstructions under sparse-view scenarios.

Submission history

From: Hyeseong Kim [view email]
[v1] Thu, 21 May 2026 07:07:37 UTC (21,550 KB)
[v2] Thu, 28 May 2026 04:53:10 UTC (21,550 KB)
[v3] Wed, 22 Jul 2026 03:03:34 UTC (21,550 KB)