DIAMOND-SSS: Diffusion-Augmented Multi-View Optimization for Data-efficient SubSurface Scattering
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915736954339328 |
|---|---|
| author | Figueroa-Araneda, Guillermo Jimenez, Iris Diana Hofherr, Florian Ko, Manny Andrade-Loarca, Hector Cremers, Daniel |
| author_facet | Figueroa-Araneda, Guillermo Jimenez, Iris Diana Hofherr, Florian Ko, Manny Andrade-Loarca, Hector Cremers, Daniel |
| contents | Subsurface scattering (SSS) gives translucent materials -- such as wax, jade, marble, and skin -- their characteristic soft shadows, color bleeding, and diffuse glow. Modeling these effects in neural rendering remains challenging due to complex light transport and the need for densely captured multi-view, multi-light datasets (often more than 100 views and 112 OLATs).
We present DIAMOND-SSS, a data-efficient framework for high-fidelity translucent reconstruction from extremely sparse supervision -- even as few as ten images. We fine-tune diffusion models for novel-view synthesis and relighting, conditioned on estimated geometry and trained on less than 7 percent of the dataset, producing photorealistic augmentations that can replace up to 95 percent of missing captures. To stabilize reconstruction under sparse or synthetic supervision, we introduce illumination-independent geometric priors: a multi-view silhouette consistency loss and a multi-view depth consistency loss.
Across all sparsity regimes, DIAMOND-SSS achieves state-of-the-art quality in relightable Gaussian rendering, reducing real capture requirements by up to 90 percent compared to SSS-3DGS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_12020 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DIAMOND-SSS: Diffusion-Augmented Multi-View Optimization for Data-efficient SubSurface Scattering Figueroa-Araneda, Guillermo Jimenez, Iris Diana Hofherr, Florian Ko, Manny Andrade-Loarca, Hector Cremers, Daniel Computer Vision and Pattern Recognition Subsurface scattering (SSS) gives translucent materials -- such as wax, jade, marble, and skin -- their characteristic soft shadows, color bleeding, and diffuse glow. Modeling these effects in neural rendering remains challenging due to complex light transport and the need for densely captured multi-view, multi-light datasets (often more than 100 views and 112 OLATs). We present DIAMOND-SSS, a data-efficient framework for high-fidelity translucent reconstruction from extremely sparse supervision -- even as few as ten images. We fine-tune diffusion models for novel-view synthesis and relighting, conditioned on estimated geometry and trained on less than 7 percent of the dataset, producing photorealistic augmentations that can replace up to 95 percent of missing captures. To stabilize reconstruction under sparse or synthetic supervision, we introduce illumination-independent geometric priors: a multi-view silhouette consistency loss and a multi-view depth consistency loss. Across all sparsity regimes, DIAMOND-SSS achieves state-of-the-art quality in relightable Gaussian rendering, reducing real capture requirements by up to 90 percent compared to SSS-3DGS. |
| title | DIAMOND-SSS: Diffusion-Augmented Multi-View Optimization for Data-efficient SubSurface Scattering |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.12020 |