# DIAMOND-SSS: Diffusion-Augmented Multi-View Optimization for Data-efficient SubSurface Scattering

arXiv: https://arxiv.org/abs/2601.12020 · HTML: https://mlsanity.com/paper/2601.12020/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 46 references that are in the map.
- First posted on arXiv 17 Jan 2026.

## Builds on (most cited first, 25 of 25)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Zero-1-to-3: Zero-shot One Image to 3D Object](https://mlsanity.com/paper/2303.11328/) (2023)
- [Depth Anything V2](https://mlsanity.com/paper/2406.09414/) (2024)
- [MVDream: Multi-view Diffusion for 3D Generation](https://mlsanity.com/paper/2308.16512/) (2023)
- [SyncDreamer: Generating Multiview-consistent Images from a Single-view Image](https://mlsanity.com/paper/2309.03453/) (2023)
- [Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model](https://mlsanity.com/paper/2310.15110/) (2023)
- [NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesis](https://mlsanity.com/paper/2012.03927/) (2020)
- [NeRD: Neural Reflectance Decomposition from Image Collections](https://mlsanity.com/paper/2012.03918/) (2020)
- [CAT3D: Create Anything in 3D with Multi-View Diffusion Models](https://mlsanity.com/paper/2405.10314/) (2024)
- [PhySG: Inverse Rendering with Spherical Gaussians for Physics-based Material Editing and Relighting](https://mlsanity.com/paper/2104.00674/) (2021)
- [Modeling Indirect Illumination for Inverse Rendering](https://mlsanity.com/paper/2204.06837/) (2022)
- [Relightable 3D Gaussians: Realistic Point Cloud Relighting with BRDF Decomposition and Ray Tracing](https://mlsanity.com/paper/2311.16043/) (2023)
- [IRON: Inverse Rendering by Optimizing Neural SDFs and Materials from Photometric Images](https://mlsanity.com/paper/2204.02232/) (2022)
- [Free3D: Consistent Novel View Synthesis without 3D Representation](https://mlsanity.com/paper/2312.04551/) (2023)
- [Shadows Don't Lie and Lines Can't Bend! Generative Models don't know Projective Geometry...for now](https://mlsanity.com/paper/2311.17138/) (2023)
- [A Diffusion Approach to Radiance Field Relighting using Multi-Illumination Synthesis](https://mlsanity.com/paper/2409.08947/) (2024)
- [MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation](https://mlsanity.com/paper/2404.03656/) (2024)
- [LumiGauss: Relightable Gaussian Splatting in the Wild](https://mlsanity.com/paper/2408.04474/) (2024)
- [Learning Object-Centric Neural Scattering Functions for Free-Viewpoint Relighting and Scene Composition](https://mlsanity.com/paper/2303.06138/) (2023)
- [Inverse Rendering of Translucent Objects using Physical and Neural Renderers](https://mlsanity.com/paper/2305.08336/) (2023)
- [SupeRVol: Super-Resolution Shape and Reflectance Estimation in Inverse Volume Rendering](https://mlsanity.com/paper/2212.04968/) (2022)
- [Fast Image-based Neural Relighting with Translucency-Reflection Modeling](https://mlsanity.com/paper/2306.09322/) (2023)
- [Sparse Views, Near Light: A Practical Paradigm for Uncalibrated Point-light Photometric Stereo](https://mlsanity.com/paper/2404.00098/) (2024)

## Abstract (arXiv)

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.