Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

TWINGS: Thin Plate Splines Warp-aligned Initialization for Sparse-View Gaussian Splatting

About

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.

Hyeseong Kim, Geonhui Son, Deukhee Lee, Dosik Hwang• 2026

Related benchmarks

TaskDatasetResultRank
Novel View SynthesisLLFF 3-view
PSNR21.49
162
Novel View SynthesisLLFF 6-view
PSNR24.97
129
Novel View SynthesisLLFF 9-view
PSNR26.23
121
Novel View SynthesisDTU 3-view
PSNR21.52
75
Novel View SynthesisDTU 6-view
PSNR25.81
68
Novel View SynthesisMip-NeRF 360 12-view
PSNR20.35
53
Novel View SynthesisDTU 9-view
PSNR28.22
41
Novel View SynthesisMip-NeRF 360 24-view
PSNR24.17
38
Showing 8 of 8 rows

Other info

Follow for update