Residual Gaussian Splatting for Ultra Sparse-View CBCT Reconstruction

TL;DR

This paper introduces Residual Gaussian Splatting, a novel method combining wavelet analysis with 3D Gaussian splatting to improve ultra sparse-view CBCT reconstruction, capturing high-frequency details while maintaining physical consistency.

Contribution

It proposes a spectrally-decoupled Gaussian representation and a spectral-spatial optimization strategy to enhance detail preservation in sparse-view CBCT imaging.

Findings

RGS captures highly refined geometric textures.

It outperforms existing neural rendering baselines.

It effectively balances artifact suppression and detail preservation.

Abstract

While 3D Gaussian splatting (3DGS) offers explicit and efficient scene representations for cone-beam computed tomography reconstruction, conventional photometric optimization inherently suffers from spectral bias under ultra sparse-view conditions, leading to over-smoothing and a loss of high-frequency anatomical details. Since wavelet transforms provide rich high-frequency information and have been widely utilized to enhance sparse reconstruction, this work integrates wavelet multi-resolution analysis with 3DGS. To circumvent the mathematical mismatch between the strict non-negativity of physical X-ray attenuation and the bipolar nature of high-frequency wavelet coefficients, we propose Residual Gaussian Splatting (RGS). Methodologically, we introduce a spectrally-decoupled Gaussian representation that stratifies the volumetric field into a geometric base component and a residual detail component. This decomposition systematically transforms explicit high-frequency fitting into a physically consistent, implicit residual compensation task. Furthermore, we devise a spectral-spatial collaborative optimization strategy to coordinate the interplay between geometric anchoring and texture refinement, effectively preventing spectral crosstalk. Extensive experiments on clinical datasets demonstrate that RGS enables the reconstructed images to capture highly refined geometric textures. It successfully resolves the trade-off between artifact suppression and detail preservation, yielding superior visual fidelity in complex trabecular and vascular structures compared to existing neural rendering baselines.