What is homeworld?

kairos-homeworld/homeworld — explained in plain English

Analysis updated 2026-06-24

★ 28Audience · researcherComplexity · 4/5Setup · hard

Mindmap

mindmap
  root((repo))
    What it does
      Floor plan to 3D scene
      Furniture placement
      Controllable generation
    Status
      Paper published 2026
      No code yet
      Future release planned
    Outputs
      Interactive scenes
      Full home environments
    Audience
      Researchers
      3D scene designers

mindmap root((repo)) What it does Floor plan to 3D scene Furniture placement Controllable generation Status Paper published 2026 No code yet Future release planned Outputs Interactive scenes Full home environments Audience Researchers 3D scene designers

Click or tap to explore — scroll the page freely

What do people build with it?

USE CASE 1

Generate a fully furnished 3D home scene from a floor plan input once the inference pipeline is released.

USE CASE 2

Read the arXiv paper to understand the controllable generation approach for 3D indoor environments.

USE CASE 3

Follow the repository to be notified when training code and datasets are published.

How does it compare?

	kairos-homeworld/homeworld	audemodo/audemodo-responsive-keepalive	ayushap18/pokefolders
Stars	28	28	28
Language	—	TypeScript	Swift
Setup difficulty	hard	easy	moderate
Complexity	4/5	2/5	2/5
Audience	researcher	developer	designer

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · hard Time to first run · 1day+

No code or datasets are currently available, training code and inference pipeline are planned for future release.

No license information is provided in this repository.

So what is it?

HomeWorld is a research project focused on generating fully furnished 3D home environments from a floorplan. You give the system a floor plan and it produces a complete, interactive scene for an entire home, including furniture placement and the ability to interact with objects throughout. The goal is controllable generation, meaning the output can be steered rather than being purely random. The project accompanies an academic paper published on arXiv in 2026. The repository currently contains no code or datasets. The authors note that training code, an inference pipeline, and datasets will be released here in the future. For now, the repository serves as a placeholder pointing to the project webpage and the arXiv paper for anyone who wants to read the technical details of the approach.

Copy-paste prompts

Prompt 1

The HomeWorld paper describes controllable 3D home generation from floor plans, summarize the key technical approach and what makes it different from prior work.

Prompt 2

Once HomeWorld releases its inference code, how would I feed a custom floor plan image into the pipeline to get a furnished 3D scene?

Prompt 3

What evaluation metrics are typically used to measure quality and controllability in 3D indoor scene generation research?

Frequently asked questions