Hierarchical Detail Enhancing Mesh-Based Shape Generation with 3D Generative Adversarial Network
This addresses the need for mesh-based shape generation in fields like industrial design and gaming, where current methods often rely on voxel or point-cloud representations, making it a domain-specific incremental advancement.
The study tackled the problem of automatic mesh-based shape generation for design and graphics communities by proposing a novel framework that uses signed distance function representation to generate detail-preserving 3D surface meshes with a deep learning approach.
Automatic mesh-based shape generation is of great interest across a wide range of disciplines, from industrial design to gaming, computer graphics and various other forms of digital art. While most traditional methods focus on primitive based model generation, advances in deep learning made it possible to learn 3-dimensional geometric shape representations in an end-to-end manner. However, most current deep learning based frameworks focus on the representation and generation of voxel and point-cloud based shapes, making it not directly applicable to design and graphics communities. This study addresses the needs for automatic generation of mesh-based geometries, and propose a novel framework that utilizes signed distance function representation that generates detail preserving three-dimensional surface mesh by a deep learning based approach.