AltNeRF: Learning Robust Neural Radiance Field via Alternating Depth-Pose Optimization
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911782413533184 |
|---|---|
| author | Wang, Kun Yan, Zhiqiang Tian, Huang Zhang, Zhenyu Li, Xiang Li, Jun Yang, Jian |
| author_facet | Wang, Kun Yan, Zhiqiang Tian, Huang Zhang, Zhenyu Li, Xiang Li, Jun Yang, Jian |
| contents | Neural Radiance Fields (NeRF) have shown promise in generating realistic novel views from sparse scene images. However, existing NeRF approaches often encounter challenges due to the lack of explicit 3D supervision and imprecise camera poses, resulting in suboptimal outcomes. To tackle these issues, we propose AltNeRF -- a novel framework designed to create resilient NeRF representations using self-supervised monocular depth estimation (SMDE) from monocular videos, without relying on known camera poses. SMDE in AltNeRF masterfully learns depth and pose priors to regulate NeRF training. The depth prior enriches NeRF's capacity for precise scene geometry depiction, while the pose prior provides a robust starting point for subsequent pose refinement. Moreover, we introduce an alternating algorithm that harmoniously melds NeRF outputs into SMDE through a consistence-driven mechanism, thus enhancing the integrity of depth priors. This alternation empowers AltNeRF to progressively refine NeRF representations, yielding the synthesis of realistic novel views. Extensive experiments showcase the compelling capabilities of AltNeRF in generating high-fidelity and robust novel views that closely resemble reality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2308_10001 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | AltNeRF: Learning Robust Neural Radiance Field via Alternating Depth-Pose Optimization Wang, Kun Yan, Zhiqiang Tian, Huang Zhang, Zhenyu Li, Xiang Li, Jun Yang, Jian Computer Vision and Pattern Recognition Neural Radiance Fields (NeRF) have shown promise in generating realistic novel views from sparse scene images. However, existing NeRF approaches often encounter challenges due to the lack of explicit 3D supervision and imprecise camera poses, resulting in suboptimal outcomes. To tackle these issues, we propose AltNeRF -- a novel framework designed to create resilient NeRF representations using self-supervised monocular depth estimation (SMDE) from monocular videos, without relying on known camera poses. SMDE in AltNeRF masterfully learns depth and pose priors to regulate NeRF training. The depth prior enriches NeRF's capacity for precise scene geometry depiction, while the pose prior provides a robust starting point for subsequent pose refinement. Moreover, we introduce an alternating algorithm that harmoniously melds NeRF outputs into SMDE through a consistence-driven mechanism, thus enhancing the integrity of depth priors. This alternation empowers AltNeRF to progressively refine NeRF representations, yielding the synthesis of realistic novel views. Extensive experiments showcase the compelling capabilities of AltNeRF in generating high-fidelity and robust novel views that closely resemble reality. |
| title | AltNeRF: Learning Robust Neural Radiance Field via Alternating Depth-Pose Optimization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2308.10001 |