Seed3D 1.0: From Images to High-Fidelity Simulation-Ready 3D Assets

2025-10-22

AI Summary

Embodied AI training is hindered by a fundamental trade-off in world simulators between content diversity and physical fidelity: video-based generation lacks real-time physical feedback, while physics engines rely on labor-intensive manual modeling and suffer from poor scalability. This paper introduces the first end-to-end framework that synthesizes simulation-ready 3D assets from a single input image—yielding high-fidelity geometry, texture-geometry alignment, and physically consistent material parameters directly compatible with mainstream physics engines (e.g., PyBullet, MuJoCo). Our approach unifies foundation models, neural 3D reconstruction, inverse rendering, and physics-aware material parameter estimation. Experiments demonstrate that the generated assets enable photorealistic robotic manipulation simulation and large-scale scene construction, significantly improving both content generation efficiency and diversity without compromising physical plausibility. The code and models are publicly released.

Technology Category

• Intelligent Robots: Embodied AI
• Computer Vision: Low Level & Physics-based Vision
• Humans and AI: Game Design — Procedural Content Generation & Storytelling

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Developing embodied AI agents requires scalable training environments that balance content diversity with physics accuracy. World simulators provide such environments but face distinct limitations: video-based methods generate diverse content but lack real-time physics feedback for interactive learning, while physics-based engines provide accurate dynamics but face scalability limitations from costly manual asset creation. We present Seed3D 1.0, a foundation model that generates simulation-ready 3D assets from single images, addressing the scalability challenge while maintaining physics rigor. Unlike existing 3D generation models, our system produces assets with accurate geometry, well-aligned textures, and realistic physically-based materials. These assets can be directly integrated into physics engines with minimal configuration, enabling deployment in robotic manipulation and simulation training. Beyond individual objects, the system scales to complete scene generation through assembling objects into coherent environments. By enabling scalable simulation-ready content creation, Seed3D 1.0 provides a foundation for advancing physics-based world simulators. Seed3D 1.0 is now available on https://console.volcengine.com/ark/region:ark+cn-beijing/experience/vision?modelId=doubao-seed3d-1-0-250928&tab=Gen3D

Problem

• Generates simulation-ready 3D assets from single images
• Addresses scalability limitations of manual physics asset creation
• Produces assets with accurate geometry and physics-compatible materials

Innovation

• Generates simulation-ready 3D assets from single images
• Produces assets with accurate geometry and aligned textures
• Enables direct integration into physics engines with minimal configuration