CSG-Based ML-Supported 3D Translation of Sketches into Game Assets

Published in The Visual Computer, 2025

The CSG-Based ML-Supported 3D Translation paper addresses the significant challenge of translating hand-drawn 2D sketches into complex 3D models efficiently. Leveraging the precision of Constructive Solid Geometry (CSG) and machine learning advancements, the research presents a system that automates this transformation, providing real-time visualization and enabling intuitive design workflows.

Key features of this research include:

  • Utilizing CSG to combine basic geometric shapes into detailed 3D structures.
  • Developing machine learning methods for accurate sketch recognition and vertex analysis.
  • Introducing a user-friendly web interface for live 3D sketch rendering.

Applications of this system span architecture, game design, and prototyping, offering a 40% improvement in modeling automation and a 20% reduction in texture rendering time. The system democratizes access to advanced 3D modeling tools, fostering creativity and enhancing productivity.

For more information, visit the related CSG-Based ML-Supported 3D Translation Research page or the dedicated paper website.

This work will be published in The Visual Computer and marks a significant step forward in sketch-based 3D modeling research.

Recommended citation: Chen, Y., Liu, Y., & Kayar-Ceylan, G. (2025). "CSG-based ML-supported 3D translation of sketches into game assets for game designers." Visual Computer. https://doi.org/10.1007/s00371-024-03758-9
Download Paper