explaingit

telehuman/gn0

23PythonAudience · researcherComplexity · 4/5Setup · hard

TLDR

A research framework for evaluating AI agents that navigate realistic indoor 3D environments by following spoken or written directions, using Gaussian Splatting reconstructions of real spaces as the test environments.

Mindmap

mindmap
  root((gn0))
    What it does
      Navigation evaluation
      Indoor scene traversal
      Instruction following
    Components
      GN-Matrix dataset
      GN-Bench simulator
      GN-BAE model
    Technology
      3D Gaussian Splatting
      Vision Language Navigation
      CUDA GPU required
    How to use
      Download InteriorGS scenes
      Download pretrained model
      Run evaluation script
    Metrics
      Goal distance error
      Task success rate
      Path efficiency
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Evaluate how well a navigation model follows natural language directions through realistic 3D indoor scenes using standard benchmark metrics.

USE CASE 2

Download the GN-Bench simulator and run the provided GN-BAE model to measure its goal-reaching accuracy on the InteriorGS dataset.

USE CASE 3

Use the GN-Matrix dataset as a benchmark for comparing new vision-and-language navigation models against the published baseline.

Tech stack

PythonCUDA

Getting it running

Difficulty · hard Time to first run · 1day+

Requires a GPU machine with CUDA support, plus downloading the InteriorGS scene dataset and pretrained model weights separately before running evaluation.

In plain English

GN0 is a research framework for training and testing AI agents that navigate indoor spaces by following spoken or written directions. The broader research field is called Vision-and-Language Navigation, where the goal is to build AI systems that can understand an instruction like "go to the bedroom and stand near the lamp" and then physically move through a 3D environment to carry it out. What makes this project distinct is its use of 3D Gaussian Splatting as the underlying scene representation. Rather than working with simple 3D models or pre-rendered video, 3D Gaussian Splatting is a newer technique for reconstructing real indoor spaces from photographs in a way that looks highly realistic when rendered from any viewpoint. GN0 uses these reconstructed scenes as the environments where agents are tested. The framework has three main components. GN-Matrix is a large dataset of navigation routes through these realistic scenes, including virtual human figures. GN-Bench is the simulation environment where agents are run and evaluated. GN-BAE is a navigation model trained to actually follow instructions through these spaces. This repository specifically contains the GN-Bench evaluation workflow. A researcher can download the InteriorGS scene dataset and a pre-trained navigation model, then run a script to evaluate how well the model performs. The evaluation reports standard metrics for this field: how far the agent ends up from its target, whether it reached the goal at all, and how efficiently it traveled compared to the shortest possible path. The project is associated with an academic paper published in 2026 and is intended for researchers working on embodied AI, robotics simulation, and language-guided navigation. Setting it up requires a GPU machine with CUDA support.

Copy-paste prompts

Prompt 1
Download the InteriorGS scene dataset and run the GN-Bench evaluation on the pre-trained GN-BAE model to get navigation success rate and path efficiency metrics.
Prompt 2
How does 3D Gaussian Splatting differ from traditional 3D meshes as a way to build realistic simulation environments for navigation research?
Prompt 3
Set up the GN0 benchmark on a CUDA GPU machine and evaluate my own navigation model using the same metrics as the published GN-BAE baseline.
Prompt 4
What success rate and path length efficiency does the GN-BAE model achieve on the GN-Bench evaluation? Where do I find the published benchmark numbers in the 2026 paper?
Open on GitHub → Explain another repo

← telehuman on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.