InfiniDream: Open-Ended Photorealistic 3D Worlds Through Procedural Generation

Hugging Face Daily Papers Papers
procedural-generation 3d-worlds vision-language-models text-to-world scene-synthesis photorealism

Summary

InfiniDream 是一个由视觉语言模型驱动的框架，通过符号化约束资产生成与确定性场景放置，实现开放、可编辑、照片级真实的 3D 世界程序化生成，并在物体与场景生成任务上验证了其效果。

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. Existing VLM-guided procedural methods remain largely constrained by predefined factories and APIs, limiting their ability to create novel asset types. We introduce InfiniDream, a VLM-driven framework for open-ended procedural 3D world generation. First, we propose Symbolically Constrained Asset Generation, which leverages existing procedural algorithms to create novel, photorealistic assets while preserving editability and seed-based diversity. Second, we develop an end-to-end text-to-world pipeline that combines semantic scene planning with Deterministic Frustum-Guided Grid Placement to construct large-scale 3D worlds with precise object placement, procedural population, and animation. Extensive evaluations on 100 object and 20 scene generation tasks demonstrate improvements across four perceptual metrics. Diversity analysis, ablation studies, and human preference evaluations further validate the effectiveness of our framework. Project page: https://yichuanh.github.io/infinidream-page/
Original Article
View Cached Full Text

Cached at: 10/20/26, 07:35 AM

Paper page - InfiniDream: Open-Ended Photorealistic 3D Worlds Through Procedural Generation

Source: https://huggingface.co/papers/2610.26853

Abstract

InfiniDream is a vision-language model framework that generates open-ended, editable, photorealistic 3D worlds through symbolic procedural asset creation and deterministic scene placement.

Procedural generationplays 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. ExistingVLM-guided procedural methods remain largely constrained by predefined factories and APIs, limiting their ability to create novel asset types. We introduce InfiniDream, aVLM-driven framework for open-ended procedural 3D world generation. First, we proposeSymbolically Constrained Asset Generation, which leverages existing procedural algorithms to create novel, photorealistic assets while preserving editability and seed-based diversity. Second, we develop an end-to-endtext-to-world pipelinethat combinessemantic scene planningwithDeterministic Frustum-Guided Grid Placementto construct large-scale 3D worlds with precise object placement, procedural population, and animation. Extensive evaluations on 100 object and 20 scene generation tasks demonstrate improvements across four perceptual metrics. Diversity analysis, ablation studies, and human preference evaluations further validate the effectiveness of our framework. Project page: https://yichuanh.github.io/infinidream-page/

View arXiv pageView PDFProject pageGitHub3Add to collection

Get this paper in your agent:

hf papers read 2610\.26853

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.26853 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.26853 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.26853 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

DreamX-World 1.0: A General-Purpose Interactive World Model

Hugging Face Daily Papers

DreamX-World 1.0 is a general-purpose interactive text/image-to-video world model that supports camera navigation, scene persistence, and promptable events across multiple domains, using novel techniques like E-PRoPE, causal forcing, and memory-conditioned scene persistence to achieve controllable long-horizon generation.

Generative World Simulation with DAO and JING

Hugging Face Daily Papers

提出 Generative World Simulation 框架，将持久共享世界状态引擎 DAO 与第一人称交互体验模型 JING 耦合，通过连续交互循环生成可随用户动作演化并保留历史的沉浸式视听体验。

OuroWorld: Bringing Any 3D World Alive as Diverse, Endlessly Looping 3D Cinemagraphs

Hugging Face Daily Papers

OuroWorld 是一个无掩码框架，可将任意静态 3D Gaussian Splatting 场景转化为从任意视角无缝循环的 3D 动态摄影（cinemagraph），利用视觉语言模型引导视频模型生成参考视频，并提出 Inconsistency-Robust Periodic 4DGS 与 Grounded Drift Field 来保证循环一致性并吸收跨视角不一致，在用户研究中以 70.8%-99.0% 的胜率超越所有基线。