Seed3D 2.0: Advancing High-Fidelity Simulation-Ready 3D Content Generation

Seed3D Team

Project Page

• We present Seed3D 2.0, an advanced 3D generation system that improves fidelity, material quality, and simulation readiness over Seed3D 1.0. It combines coarse-to-fine geometry generation, unified PBR material synthesis, and simulation-ready scene and part-level modeling for high-quality interactive 3D assets.

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

Seed3D Team

Project Page

• 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.

Threestudio: A unified framework for 3D content generation

Threestudio Team

Project Page

• We introduce threestudio, an open-source, unified, and modular framework specifically designed for 3D content generation. This framework extends diffusion-based 2D image generation models to 3D generation guidance while incorporating conditions such as text and images. We delineate the modular architecture and design of each component within threestudio.

Pygmalion: Bridging Reconstruction and Generation in Sparse Voxel-based 3D Modeling

Guan Luo, Jing Lin, Xuanyu Yi, Jiahang Liu, Song-Hai Zhang, Jianfeng Zhang

Project Page

• We introduce Pygmalion, a sparse voxel autoencoding framework that bridges high-fidelity reconstruction and robust 3D generation. Its coupled SDF parameterization turns decoding perturbations into smooth geometric changes, while sign correction, geometry supervision, and rendering refinement help produce complete surfaces with sharp features and fine details.

TopoMesh: High-Fidelity Mesh Autoencoding via Topological Unification

Guan Luo, Xiu Li, Rui Chen, Xuanyu Yi, Jing Lin, Chia-Hao Chen, Jiahang Liu, Song-Hai Zhang, Jianfeng Zhang

Project Page

• We introduce TopoMesh, a sparse voxel-based VAE that unifies both GT and predicted meshes under a shared Dual Marching Cubes topological framework. This establishes explicit correspondences at the vertex and face level, allowing us to derive explicit mesh-level supervision signals for topology, vertex positions, and face orientations with clear gradients.

LaFiTe: A Generative Latent Field for 3D Native Texturing

Chia-Hao Chen, Zi-Xin Zou, Yan-Pei Cao, Ze Yuan, Guan Luo, Xiaojuan Qi, Ding Liang, Song-Hai Zhang, Yuan-Chen Guo

Project Page

• We introduce LaFiTe, a framework that addresses this challenge by learning to generate textures as a 3D generative sparse latent color field.

MS3D: High-Quality 3D Generation via Multi-Scale Representation Modeling

Guan Luo, Jianfeng Zhang

Project Page

• We introduce MS3D, a novel multi-scale 3D reconstruction framework. At its core, we introduce a hierarchical structured latent representation for multi-scale modeling, coupled with a multi-scale feature extraction and integration mechanism, which enables progressive reconstruction, effectively decomposing the complex task of detailed geometry reconstruction into a sequence of easier steps.

3D Gaussian Editing With A Single Image

Guan Luo, Tian-Xing Xu, Ying-Tian Liu, Xiao-Xiong Fan, Fang-Lue Zhang, Song-Hai Zhang

Project Page

• We introduce a novel single-image-driven 3D scene editing approach based on 3D Gaussian Splatting, enabling intuitive manipulation via directly editing the content on a 2D image plane. Our method learns to optimize the 3D Gaussians to align with an edited version of the image rendered from a user-specified viewpoint of the original scene.