This repository is a curated list of academic papers on 3D scene generation, targeting researchers and practitioners in computer graphics and AI. It provides a structured overview of methods, datasets, and applications, aiming to serve as a comprehensive resource for the field.

How It Works

The repository categorizes 3D scene generation techniques into a hierarchical taxonomy, including procedural, rule-based, optimization-based, LLM-based, and neural generation methods. It also lists relevant datasets and downstream applications like 3D scene editing, embodied AI, and autonomous driving. The content is presented in a tabular format, detailing publication year, venue, paper title, and links to projects or GitHub repositories where available.

Quick Start & Requirements

This repository is a curated list of papers and does not have direct installation or execution requirements. It serves as a reference guide.

Highlighted Details

• Comprehensive coverage of over 300 recent studies on 3D scene generation.
• Hierarchical taxonomy of methods, from traditional procedural generation to modern LLM-based approaches.
• Extensive lists of datasets (indoor, natural, urban) and applications.
• Includes links to associated projects and GitHub repositories for many papers.

Maintenance & Community

The repository is maintained by hzxie and welcomes community contributions through issues and pull requests for new resources, methodology improvements, or corrections.

Licensing & Compatibility

The repository itself is not software and thus does not have a software license. The licenses of linked projects or papers would need to be checked individually.

Limitations & Caveats

As a curated list, the repository's content is limited to publicly available academic papers and their associated resources. The availability and maintenance of linked GitHub repositories are external to this project.