MeTTA: Single-View to 3D Textured Mesh Reconstruction with Test-Time Adaptation
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_ | 1866911997740711936 |
|---|---|
| author | Yu-Ji, Kim Ha, Hyunwoo Youwang, Kim Surh, Jaeheung Ha, Hyowon Oh, Tae-Hyun |
| author_facet | Yu-Ji, Kim Ha, Hyunwoo Youwang, Kim Surh, Jaeheung Ha, Hyowon Oh, Tae-Hyun |
| contents | Reconstructing 3D from a single view image is a long-standing challenge. One of the popular approaches to tackle this problem is learning-based methods, but dealing with the test cases unfamiliar with training data (Out-of-distribution; OoD) introduces an additional challenge. To adapt for unseen samples in test time, we propose MeTTA, a test-time adaptation (TTA) exploiting generative prior. We design joint optimization of 3D geometry, appearance, and pose to handle OoD cases with only a single view image. However, the alignment between the reference image and the 3D shape via the estimated viewpoint could be erroneous, which leads to ambiguity. To address this ambiguity, we carefully design learnable virtual cameras and their self-calibration. In our experiments, we demonstrate that MeTTA effectively deals with OoD scenarios at failure cases of existing learning-based 3D reconstruction models and enables obtaining a realistic appearance with physically based rendering (PBR) textures. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_11465 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MeTTA: Single-View to 3D Textured Mesh Reconstruction with Test-Time Adaptation Yu-Ji, Kim Ha, Hyunwoo Youwang, Kim Surh, Jaeheung Ha, Hyowon Oh, Tae-Hyun Computer Vision and Pattern Recognition Reconstructing 3D from a single view image is a long-standing challenge. One of the popular approaches to tackle this problem is learning-based methods, but dealing with the test cases unfamiliar with training data (Out-of-distribution; OoD) introduces an additional challenge. To adapt for unseen samples in test time, we propose MeTTA, a test-time adaptation (TTA) exploiting generative prior. We design joint optimization of 3D geometry, appearance, and pose to handle OoD cases with only a single view image. However, the alignment between the reference image and the 3D shape via the estimated viewpoint could be erroneous, which leads to ambiguity. To address this ambiguity, we carefully design learnable virtual cameras and their self-calibration. In our experiments, we demonstrate that MeTTA effectively deals with OoD scenarios at failure cases of existing learning-based 3D reconstruction models and enables obtaining a realistic appearance with physically based rendering (PBR) textures. |
| title | MeTTA: Single-View to 3D Textured Mesh Reconstruction with Test-Time Adaptation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2408.11465 |