Stars
Code for "ATISS: Autoregressive Transformers for Indoor Scene Synthesis", NeurIPS 2021
BoxFusion: Reconstruction-Free Open-Vocabulary 3D Object Detection via Real-Time Multi-View Box Fusion
We present SceneFactor, a diffusion-based approach for large-scale 3D scene generation that enables controllable generation and effortless editing.
PhyRecon: Physically Plausible Neural Scene Reconstruction
HoloPart: Generative 3D Part Amodal Segmentation
This repository records useful knowledge and tips in the field of 3D vision. Continuously updates... Welcome to 🌟
PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers
CLR-Wire: Towards Continuous Latent Representations for 3D Curve Wireframe Generation (ACM SIGGRAPH 2025)
😎 A list of awesome Computer-Aided Design (CAD) papers
Symbol as Points: Panoptic Symbol Spotting via Point-based Representation. ICLR 2024
Materials for the Hugging Face Diffusion Models Course
Code for the TNNLS 2023 paper "STEdge: Self-training Edge Detection with Multi-layer Teaching and Regularization".
Code for the TIP 2023 paper "Delving into Crispness: Guided Label Refinement for Crisp Edge Detection".
PlaneRecTR++: Unified Query Learning for Joint 3D Planar Reconstruction and Pose Estimation
[CVPR 2025] OnlineAnySeg: Online Zero-Shot 3D Segmentation by Visual Foundation Model Guided 2D Mask Merging
Scene Graph Disentanglement and Composition for Generalizable Complex Image Generation
程序员在家做饭方法指南。Programmer's guide about how to cook at home (Simplified Chinese only).
[NeurIPS 2024] SceneCraft: Layout-Guided 3D Scene Generation
[RSS2024] Official implementation of "Hierarchical Open-Vocabulary 3D Scene Graphs for Language-Grounded Robot Navigation"
Janus-Series: Unified Multimodal Understanding and Generation Models
[RA-L 2025] Dynamic Open-Vocabulary 3D Scene Graphs for Long-term Language-Guided Mobile Manipulation
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch and FLAX.
Official PyTorch Implementation of "Learning Self-supervised Part-aware 3D Hybrid Representations of 2D Gaussians and Superquadrics"
[ICLR 2025] EditRoom: LLM-parameterized Graph Diffusion for Composable 3D Room Layout Editing
[NeurIPS'2024] Zero-Shot Scene Reconstruction from Single Images with Deep Prior Assembly