inspatio/querysplat

repo for QuerySplat

What it solves

QuerySplat addresses the challenge of predicting 3D Gaussian Splatting (3DGS) representations from images by decoupling the geometry and appearance representations, allowing for more accurate 3D scene reconstruction.

How it works

The system uses a feed-forward model to predict 3D Gaussians and render them. It integrates with VGGT-Omega for image encoding, camera prediction, and depth prediction. The pipeline includes custom-image preprocessing and offers an optional Test-Time Optimization (TTO) step to further refine the predicted 3D scene based on reconstruction objectives like LPIPS.

Who it’s for

This project is designed for researchers and developers working in 3D computer vision, specifically those focused on 3D Gaussian Splatting and neural scene representation.

Highlights

  • Decouples geometry and appearance for better 3DGS prediction.
  • Supports both fast feed-forward inference and refined test-time optimization (TTO).
  • Integrates with VGGT-Omega for robust camera and depth estimation.
  • Provides tools for exporting Gaussian PLY files and analyzing opacity and scale distributions.

Related

  • Project
    Linketic/CityGaussian

    A framework for high-quality, large-scale 3D scene reconstruction using Gaussian Splatting, featuring multi-GPU support and geometrically accurate rendering.

  • Project
    cvg/resplat

    ReSplat is a feed-forward recurrent model for 3D Gaussian splatting that iteratively refines scene representations using rendering error as feedback.

  • Dispatch
    Introduction to 3D Gaussian Splatting

    3D Gaussian Splatting is a rasterization technique that enables real-time rendering of photorealistic 3D scenes learned from a small set of images.

  • Project
    zju3dv/InfiniSplat

    InfiniSplat is a framework for large-baseline monocular view synthesis that reconstructs 3D Gaussian Splatting scenes from a single RGB image or an RGB-depth pair.

  • Dispatch
    Gaussian Point Splatting: Scaling Novel View Synthesis to Hundreds of Millions of Gaussians

    Gaussian Point Splatting is a stochastic rendering method presented at SIGGRAPH 2026 that enables real-time rendering of hundreds of millions of Gaussians using GPU atomics and parallel programming primitives.