InfiniDream generates editable photorealistic 3D worlds with VLMs
Original title: InfiniDream: Open-Ended Photorealistic 3D Worlds Through Procedural Generation
Evaluations on 100 object and 20 scene tasks report gains on four perceptual metrics.
InfiniDream is a VLM-driven framework for open-ended procedural 3D world generation. Symbolically Constrained Asset Generation uses existing procedural algorithms to make novel photorealistic assets while keeping editability and seed-based diversity. A text-to-world pipeline pairs semantic scene planning with Deterministic Frustum-Guided Grid Placement. Diversity analysis, ablations, and human preference evaluations are also reported.
Unconfirmed: Metric gains are author-reported; exact score values are not given in the abstract.
- The method aims to create asset types beyond predefined factories and APIs.
- Placement covers large-scale worlds, procedural population, and animation.
- A public project page documents the framework.
From the original article
1 min read at huggingface.co
Procedural generation plays a key role in synthetic data creation and game development, offering greater controllability, editability, and diversity than static mesh representations. However, VLM-based code generation often produces relatively simple 3D assets, while sophisticated photorealistic procedural systems require extensive manual development.
By Yi-Chuan Huang, Hao-Jen Chien, Zheng-Hui Huang · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.