Search papers, labs, and topics across Lattice.
This paper introduces DAV-GSWT, a framework for data-efficient Gaussian Splatting Wang Tile generation that reduces reliance on densely sampled exemplar reconstructions. It uses diffusion priors and active view sampling, guided by a hierarchical uncertainty quantification mechanism, to identify informative viewpoints and hallucinate missing details for seamless tile transitions. Experiments demonstrate that DAV-GSWT maintains visual integrity and interactive performance for large-scale environments while significantly reducing the required data volume.
Generate expansive, high-fidelity 3D environments with Gaussian Splatting using significantly less data by intelligently sampling viewpoints and hallucinating details with diffusion models.
The emergence of 3D Gaussian Splatting has fundamentally redefined the capabilities of photorealistic neural rendering by enabling high-throughput synthesis of complex environments. While procedural methods like Wang Tiles have recently been integrated to facilitate the generation of expansive landscapes, these systems typically remain constrained by a reliance on densely sampled exemplar reconstructions. We present DAV-GSWT, a data-efficient framework that leverages diffusion priors and active view sampling to synthesize high-fidelity Gaussian Splatting Wang Tiles from minimal input observations. By integrating a hierarchical uncertainty quantification mechanism with generative diffusion models, our approach autonomously identifies the most informative viewpoints while hallucinating missing structural details to ensure seamless tile transitions. Experimental results indicate that our system significantly reduces the required data volume while maintaining the visual integrity and interactive performance necessary for large-scale virtual environments.
