Fine Structure-Aware Sampling: A New Sampling Training Scheme for Pixel-Aligned Implicit Models in Single-View Human Reconstruction
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_ | 1866909383858847744 |
|---|---|
| author | Chan, Kennard Yanting Liu, Fayao Lin, Guosheng Foo, Chuan Sheng Lin, Weisi |
| author_facet | Chan, Kennard Yanting Liu, Fayao Lin, Guosheng Foo, Chuan Sheng Lin, Weisi |
| contents | Pixel-aligned implicit models, such as PIFu, PIFuHD, and ICON, are used for single-view clothed human reconstruction. These models need to be trained using a sampling training scheme. Existing sampling training schemes either fail to capture thin surfaces (e.g. ears, fingers) or cause noisy artefacts in reconstructed meshes. To address these problems, we introduce Fine Structured-Aware Sampling (FSS), a new sampling training scheme to train pixel-aligned implicit models for single-view human reconstruction. FSS resolves the aforementioned problems by proactively adapting to the thickness and complexity of surfaces. In addition, unlike existing sampling training schemes, FSS shows how normals of sample points can be capitalized in the training process to improve results. Lastly, to further improve the training process, FSS proposes a mesh thickness loss signal for pixel-aligned implicit models. It becomes computationally feasible to introduce this loss once a slight reworking of the pixel-aligned implicit function framework is carried out. Our results show that our methods significantly outperform SOTA methods qualitatively and quantitatively. Our code is publicly available at https://github.com/kcyt/FSS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_19197 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Fine Structure-Aware Sampling: A New Sampling Training Scheme for Pixel-Aligned Implicit Models in Single-View Human Reconstruction Chan, Kennard Yanting Liu, Fayao Lin, Guosheng Foo, Chuan Sheng Lin, Weisi Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Pixel-aligned implicit models, such as PIFu, PIFuHD, and ICON, are used for single-view clothed human reconstruction. These models need to be trained using a sampling training scheme. Existing sampling training schemes either fail to capture thin surfaces (e.g. ears, fingers) or cause noisy artefacts in reconstructed meshes. To address these problems, we introduce Fine Structured-Aware Sampling (FSS), a new sampling training scheme to train pixel-aligned implicit models for single-view human reconstruction. FSS resolves the aforementioned problems by proactively adapting to the thickness and complexity of surfaces. In addition, unlike existing sampling training schemes, FSS shows how normals of sample points can be capitalized in the training process to improve results. Lastly, to further improve the training process, FSS proposes a mesh thickness loss signal for pixel-aligned implicit models. It becomes computationally feasible to introduce this loss once a slight reworking of the pixel-aligned implicit function framework is carried out. Our results show that our methods significantly outperform SOTA methods qualitatively and quantitatively. Our code is publicly available at https://github.com/kcyt/FSS. |
| title | Fine Structure-Aware Sampling: A New Sampling Training Scheme for Pixel-Aligned Implicit Models in Single-View Human Reconstruction |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2402.19197 |