Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.05330 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912965096112128 |
|---|---|
| author | Guo, Andrew Y Malik, Anagh Tedla, SaiKiran Dai, Yutong Qin, Yiqian Salehe, Zach Attal, Benjamin Nousias, Sotiris Kutulakos, Kiriakos N. Lindell, David B. |
| author_facet | Guo, Andrew Y Malik, Anagh Tedla, SaiKiran Dai, Yutong Qin, Yiqian Salehe, Zach Attal, Benjamin Nousias, Sotiris Kutulakos, Kiriakos N. Lindell, David B. |
| contents | We introduce Dark3R, a framework for structure from motion in the dark that operates directly on raw images with signal-to-noise ratios (SNRs) below $-4$ dB -- a regime where conventional feature- and learning-based methods break down. Our key insight is to adapt large-scale 3D foundation models to extreme low-light conditions through a teacher--student distillation process, enabling robust feature matching and camera pose estimation in low light. Dark3R requires no 3D supervision; it is trained solely on noisy--clean raw image pairs, which can be either captured directly or synthesized using a simple Poisson--Gaussian noise model applied to well-exposed raw images. To train and evaluate our approach, we introduce a new, exposure-bracketed dataset that includes $\sim$42,000 multi-view raw images with ground-truth 3D annotations, and we demonstrate that Dark3R achieves state-of-the-art structure from motion in the low-SNR regime. Further, we demonstrate state-of-the-art novel view synthesis in the dark using Dark3R's predicted poses and a coarse-to-fine radiance field optimization procedure. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_05330 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Dark3R: Learning Structure from Motion in the Dark Guo, Andrew Y Malik, Anagh Tedla, SaiKiran Dai, Yutong Qin, Yiqian Salehe, Zach Attal, Benjamin Nousias, Sotiris Kutulakos, Kiriakos N. Lindell, David B. Computer Vision and Pattern Recognition We introduce Dark3R, a framework for structure from motion in the dark that operates directly on raw images with signal-to-noise ratios (SNRs) below $-4$ dB -- a regime where conventional feature- and learning-based methods break down. Our key insight is to adapt large-scale 3D foundation models to extreme low-light conditions through a teacher--student distillation process, enabling robust feature matching and camera pose estimation in low light. Dark3R requires no 3D supervision; it is trained solely on noisy--clean raw image pairs, which can be either captured directly or synthesized using a simple Poisson--Gaussian noise model applied to well-exposed raw images. To train and evaluate our approach, we introduce a new, exposure-bracketed dataset that includes $\sim$42,000 multi-view raw images with ground-truth 3D annotations, and we demonstrate that Dark3R achieves state-of-the-art structure from motion in the low-SNR regime. Further, we demonstrate state-of-the-art novel view synthesis in the dark using Dark3R's predicted poses and a coarse-to-fine radiance field optimization procedure. |
| title | Dark3R: Learning Structure from Motion in the Dark |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.05330 |