GEM3D: GEnerative Medial Abstractions for 3D Shape Synthesis
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916200660860928 |
|---|---|
| author | Petrov, Dmitry Goyal, Pradyumn Thamizharasan, Vikas Kim, Vladimir G. Gadelha, Matheus Averkiou, Melinos Chaudhuri, Siddhartha Kalogerakis, Evangelos |
| author_facet | Petrov, Dmitry Goyal, Pradyumn Thamizharasan, Vikas Kim, Vladimir G. Gadelha, Matheus Averkiou, Melinos Chaudhuri, Siddhartha Kalogerakis, Evangelos |
| contents | We introduce GEM3D -- a new deep, topology-aware generative model of 3D shapes. The key ingredient of our method is a neural skeleton-based representation encoding information on both shape topology and geometry. Through a denoising diffusion probabilistic model, our method first generates skeleton-based representations following the Medial Axis Transform (MAT), then generates surfaces through a skeleton-driven neural implicit formulation. The neural implicit takes into account the topological and geometric information stored in the generated skeleton representations to yield surfaces that are more topologically and geometrically accurate compared to previous neural field formulations. We discuss applications of our method in shape synthesis and point cloud reconstruction tasks, and evaluate our method both qualitatively and quantitatively. We demonstrate significantly more faithful surface reconstruction and diverse shape generation results compared to the state-of-the-art, also involving challenging scenarios of reconstructing and synthesizing structurally complex, high-genus shape surfaces from Thingi10K and ShapeNet. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_16994 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GEM3D: GEnerative Medial Abstractions for 3D Shape Synthesis Petrov, Dmitry Goyal, Pradyumn Thamizharasan, Vikas Kim, Vladimir G. Gadelha, Matheus Averkiou, Melinos Chaudhuri, Siddhartha Kalogerakis, Evangelos Computer Vision and Pattern Recognition Artificial Intelligence Graphics Machine Learning We introduce GEM3D -- a new deep, topology-aware generative model of 3D shapes. The key ingredient of our method is a neural skeleton-based representation encoding information on both shape topology and geometry. Through a denoising diffusion probabilistic model, our method first generates skeleton-based representations following the Medial Axis Transform (MAT), then generates surfaces through a skeleton-driven neural implicit formulation. The neural implicit takes into account the topological and geometric information stored in the generated skeleton representations to yield surfaces that are more topologically and geometrically accurate compared to previous neural field formulations. We discuss applications of our method in shape synthesis and point cloud reconstruction tasks, and evaluate our method both qualitatively and quantitatively. We demonstrate significantly more faithful surface reconstruction and diverse shape generation results compared to the state-of-the-art, also involving challenging scenarios of reconstructing and synthesizing structurally complex, high-genus shape surfaces from Thingi10K and ShapeNet. |
| title | GEM3D: GEnerative Medial Abstractions for 3D Shape Synthesis |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Graphics Machine Learning |
| url | https://arxiv.org/abs/2402.16994 |