GS$^{2}$CI: Robust Gaussian Splatting For Snapshot Compressive Imaging via Large Vision Model Priors

2026-08-13

AI Summary

This work addresses the challenges of 3D reconstruction from single-shot snapshot compressive imaging (SCI), which suffers from severe information loss, limited viewpoints, and computationally intensive optimization. The authors propose a novel approach that integrates 3D Gaussian Splatting with priors from large-scale vision foundation models, enabling efficient reconstruction through measurement-driven initialization, SCI-aware Gaussian optimization, and pseudo-view supervision. A key innovation is the Opacity-Guided Splitting and Growth Regulation strategy, which leverages local opacity statistics to guide Gaussian splitting and growth while enforcing explicit densification constraints, substantially enhancing representation stability and robustness. Experimental results demonstrate that the method achieves state-of-the-art overall performance across multiple benchmarks, offering high-fidelity reconstruction, strong viewpoint robustness, and computational efficiency.

Abstract

Snapshot Compressive Imaging (SCI) offers an efficient solution for high-speed video acquisition and, under exposure-time camera--scene relative motion, multi-view scene capture by compressing temporal or spatial information into a single 2D measurement. While recent studies have explored SCI for 3D scene reconstruction, existing methods struggle with significant challenges due to information loss, limited viewpoint diversity, and the computational burden of jointly optimizing 3D representations and camera poses. In this work, we propose a novel framework that reconstructs high-quality 3D scenes from a single SCI measurement by leveraging 3D Gaussian Splatting (3DGS) and the powerful priors of large-scale vision foundation models (VFMs). Our primary reconstruction combines measurement-derived 3D VFM initialization with SCI-aware Gaussian optimization. After coarse-stage convergence, an auxiliary 2D VFM provides pseudo-view supervision at synthesized viewpoints for local appearance refinement. To further address the instability caused by ambiguous SCI supervision during 3DGS optimization, we introduce Opacity-Guided Splitting and Growth Regulation (OSGR), an SCI-specific densification strategy that augments split candidates using local opacity statistics, discourages loss-compensating opacity inflation through mean-opacity regulation, and bounds representation growth with explicit candidate-ratio and Gaussian-count constraints. Extensive experiments across multiple benchmarks demonstrate that our method achieves the strongest overall performance, combining leading reconstruction quality and robustness to viewpoint variation with competitive computational efficiency.

Problem

• Snapshot Compressive Imaging
• 3D scene reconstruction
• information loss
• viewpoint diversity
• computational burden

Innovation

• Gaussian Splatting
• Snapshot Compressive Imaging
• Vision Foundation Models
• Opacity-Guided Densification
• 3D Reconstruction