DAV-GSWT: Diffusion-Active-View Sampling for Data-Efficient Gaussian Splatting Wang Tiles

2026-02-16

AI Summary

This work addresses the limitations of traditional Wang Tile–based Gaussian splatting methods, which rely on dense sampling and suffer from low data efficiency and poor scalability to large-scale scenes. The paper introduces, for the first time, a framework that integrates diffusion-based generative priors with active view sampling. By leveraging a hierarchical uncertainty quantification mechanism, the method automatically selects the most informative observation viewpoints and synthesizes high-fidelity, structurally coherent Gaussian splatting Wang Tiles from minimal input. This enables seamless tiling while drastically improving data efficiency. The approach maintains high-quality rendering and smooth interactivity even with significantly fewer input views, making it well-suited for the efficient construction of large-scale virtual environments.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Sampling/Simulation-based Search
• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

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.

Problem

• Gaussian Splatting
• Wang Tiles
• data efficiency
• neural rendering
• 3D reconstruction

Innovation

• Diffusion Models
• Active View Sampling
• Gaussian Splatting
• Wang Tiles
• Data Efficiency