.png){kind=icon}
.png){kind=icon}
Return
Geometry-aware triplane diffusion for single shape generation with feature alignment
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cag.2025.104384.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A unified framework for single-shape 3D generation from one exemplar input. • Triplane autoencoder with spatial predictor captures fine-grained geometry. • Soft feature alignment ensures fidelity and diversity in shape synthesis. • Multi-scale diffusion improves structural coherence and training stability • The method sets a strong baseline for data-sparse 3D generative modeling.
Keywords:
3D generation
triplane autoencoder
soft feature alignment
multi-scale diffusion
shape synthesis