GS$^{2}$CI: Robust Gaussian Splatting For Snapshot Compressive Imaging via Large Vision Model Priors
2026-08-13 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors tackle the problem of creating detailed 3D scenes from a single compressed 2D video snapshot, which is hard because of lost information and limited viewpoints. They introduce a new method that uses a technique called 3D Gaussian Splatting combined with large vision models to start the 3D reconstruction and then refine it with extra 2D image information. To keep the process stable and accurate, they developed a way to control the complexity of the 3D model using opacity rules. Their experiments show their method produces better and more reliable 3D reconstructions compared to earlier approaches.
Snapshot Compressive Imaging (SCI)3D Gaussian Splatting (3DGS)Vision Foundation Models (VFMs)3D Scene ReconstructionOpacity-Guided Splitting and Growth Regulation (OSGR)Temporal CompressionMulti-view CaptureCamera Pose OptimizationPseudo-view Supervision
Authors
Yanming Yang, Chenxi Song, Ping Wang, Xin Yuan, Chi Zhang
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.