CHRIS: Clothed Human Reconstruction with Side View Consistency
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912380997337088 |
|---|---|
| author | Liu, Dong Yang, Yifan Huang, Zixiong Gao, Yuxin Tan, Mingkui |
| author_facet | Liu, Dong Yang, Yifan Huang, Zixiong Gao, Yuxin Tan, Mingkui |
| contents | Creating a realistic clothed human from a single-view RGB image is crucial for applications like mixed reality and filmmaking. Despite some progress in recent years, mainstream methods often fail to fully utilize side-view information, as the input single-view image contains front-view information only. This leads to globally unrealistic topology and local surface inconsistency in side views. To address these, we introduce Clothed Human Reconstruction with Side View Consistency, namely CHRIS, which consists of 1) A Side-View Normal Discriminator that enhances global visual reasonability by distinguishing the generated side-view normals from the ground truth ones; 2) A Multi-to-One Gradient Computation (M2O) that ensures local surface consistency. M2O calculates the gradient of a sampling point by integrating the gradients of the nearby points, effectively acting as a smooth operation. Experimental results demonstrate that CHRIS achieves state-of-the-art performance on public benchmarks and outperforms the prior work. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_12005 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CHRIS: Clothed Human Reconstruction with Side View Consistency Liu, Dong Yang, Yifan Huang, Zixiong Gao, Yuxin Tan, Mingkui Computer Vision and Pattern Recognition Artificial Intelligence Creating a realistic clothed human from a single-view RGB image is crucial for applications like mixed reality and filmmaking. Despite some progress in recent years, mainstream methods often fail to fully utilize side-view information, as the input single-view image contains front-view information only. This leads to globally unrealistic topology and local surface inconsistency in side views. To address these, we introduce Clothed Human Reconstruction with Side View Consistency, namely CHRIS, which consists of 1) A Side-View Normal Discriminator that enhances global visual reasonability by distinguishing the generated side-view normals from the ground truth ones; 2) A Multi-to-One Gradient Computation (M2O) that ensures local surface consistency. M2O calculates the gradient of a sampling point by integrating the gradients of the nearby points, effectively acting as a smooth operation. Experimental results demonstrate that CHRIS achieves state-of-the-art performance on public benchmarks and outperforms the prior work. |
| title | CHRIS: Clothed Human Reconstruction with Side View Consistency |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2505.12005 |