#
Starred repositories
Implementation of the paper Gaussian Fluids: A Grid-Free Fluid Solver based on Gaussian Spatial Representation
Towards a Generative 3D World Engine for Embodied Intelligence
Repo for baseline codes of Digital Twin Catalog project.
From Images to High-Fidelity 3D Assets with Production-Ready PBR Material
Cosmos-Reason1 models understand the physical common sense and generate appropriate embodied decisions in natural language through long chain-of-thought reasoning processes.
Official repo for paper "Sparse Representation and Construction for High-Resolution 3D Shapes Modeling".
DiffusionRenderer (Cosmos): Neural Inverse and Forward Rendering with Video Diffusion Models
Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models
Cosmos-Predict2 is a collection of general-purpose world foundation models for Physical AI that can be fine-tuned into customized world models for downstream applications.
[CVPR'25 Oral] Official implementation for "DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models"
PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers
CVPR 2025: VoxelSplat: Dynamic Gaussian Splatting as an Effective Loss for Occupancy and Flow Prediction
23
Updated Jun 8, 2025
NeuSmoke: Efficient Smoke Reconstruction and View Synthesis with Neural Transportation Fields (SIGGRAPH Asia 2024)
CVPR 2024: Language Guided Generation of 3D Embodied AI Environments.
[CVPR 2025] TexGaussian: Generating High-quality PBR Material via Octree-based 3D Gaussian Splatting
这是一个简单的技术科普教程项目，主要聚焦于解释一些有趣的，前沿的技术概念和原理。每篇文章都力求在 5 分钟内阅读完成。
Official implementation of "UniTEX: Universal High Fidelity Generative Texturing for 3D Shapes"
Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds
OctGPT: Octree-based Multiscale Autoregressive Models for 3D Shape Generation
Official Code Release for [SIGGRAPH 2025] RenderFormer: Transformer-based Neural Rendering of Triangle Meshes with Global Illumination
[NeurIPS'24] Large Spatial Model: End-to-end Unposed Images to Semantic 3D
PyTorch implementation of FractalGen https://arxiv.org/abs/2502.17437
The official repo of "Advancing high-fidelity 3D and Texture Generation with 2.5D latents"
14
Updated May 28, 2025
[NeurIPS 2024] Geometry-Aware Large Reconstruction Model for Efficient and High-Quality 3D Generation
Lightplane implements a highly memory-efficient differentiable radiance field renderer, and a module for unprojecting features from images to 3D grids.
Direct3D‑S2: Gigascale 3D Generation Made Easy with Spatial Sparse Attention
Open-Source Implementations of Multi-Modal Diffusion Models Optimized for Highest Quality and Ease of Use