3DPR: Single Image 3D Portrait Relight using Generative Priors

• URL: http://arxiv.org/abs/2510.15846v1
• Date: Fri, 17 Oct 2025 17:37:42 GMT
• Title: 3DPR: Single Image 3D Portrait Relight using Generative Priors
• Authors: Pramod Rao, Abhimitra Meka, Xilong Zhou, Gereon Fox, Mallikarjun B R, Fangneng Zhan, Tim Weyrich, Bernd Bickel, Hanspeter Pfister, Wojciech Matusik, Thabo Beeler, Mohamed Elgharib, Marc Habermann, Christian Theobalt,
• Abstract summary: 3DPR is an image-based relighting model that leverages generative priors learnt from multi-view One-Light-at-A-Time (OLAT) images.<n>We leverage the latent space of a pre-trained generative head model that provides a rich prior over face geometry learnt from in-the-wild image datasets.<n>Our reflectance network operates in the latent space of the generative head model, crucially enabling a relatively small number of lightstage images to train the reflectance model.
• Score: 101.74130664920868
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Rendering novel, relit views of a human head, given a monocular portrait image as input, is an inherently underconstrained problem. The traditional graphics solution is to explicitly decompose the input image into geometry, material and lighting via differentiable rendering; but this is constrained by the multiple assumptions and approximations of the underlying models and parameterizations of these scene components. We propose 3DPR, an image-based relighting model that leverages generative priors learnt from multi-view One-Light-at-A-Time (OLAT) images captured in a light stage. We introduce a new diverse and large-scale multi-view 4K OLAT dataset of 139 subjects to learn a high-quality prior over the distribution of high-frequency face reflectance. We leverage the latent space of a pre-trained generative head model that provides a rich prior over face geometry learnt from in-the-wild image datasets. The input portrait is first embedded in the latent manifold of such a model through an encoder-based inversion process. Then a novel triplane-based reflectance network trained on our lightstage data is used to synthesize high-fidelity OLAT images to enable image-based relighting. Our reflectance network operates in the latent space of the generative head model, crucially enabling a relatively small number of lightstage images to train the reflectance model. Combining the generated OLATs according to a given HDRI environment maps yields physically accurate environmental relighting results. Through quantitative and qualitative evaluations, we demonstrate that 3DPR outperforms previous methods, particularly in preserving identity and in capturing lighting effects such as specularities, self-shadows, and subsurface scattering. Project Page: https://vcai.mpi-inf.mpg.de/projects/3dpr/

Related papers

• Inverse Image-Based Rendering for Light Field Generation from Single Images [30.856397422416517]
  We propose a novel view synthesis method for light field generation from only single images, named inverse image-based rendering.<n>Our method reconstructs light flows in a space from image pixels, which behaves in the opposite way to image-based rendering.<n>Our neural first stores the light flow of source rays from the input image, then computes the relationships among them through cross-attention.
  arXiv  Detail & Related papers  (2025-10-23T02:12:45Z)
• RelitLRM: Generative Relightable Radiance for Large Reconstruction Models [52.672706620003765]
  We propose RelitLRM for generating high-quality Gaussian splatting representations of 3D objects under novel illuminations. Unlike prior inverse rendering methods requiring dense captures and slow optimization, RelitLRM adopts a feed-forward transformer-based model. We show our sparse-view feed-forward RelitLRM offers competitive relighting results to state-of-the-art dense-view optimization-based baselines.
  arXiv  Detail & Related papers  (2024-10-08T17:40:01Z)
• Light Field Diffusion for Single-View Novel View Synthesis [32.59286750410843]
  Single-view novel view synthesis (NVS) is important but challenging in computer vision. Recent advancements in NVS have leveraged Denoising Diffusion Probabilistic Models (DDPMs) for their exceptional ability to produce high-fidelity images. We present Light Field Diffusion (LFD), a novel conditional diffusion-based approach that transcends the conventional reliance on camera pose matrices.
  arXiv  Detail & Related papers  (2023-09-20T03:27:06Z)
• Relightify: Relightable 3D Faces from a Single Image via Diffusion Models [86.3927548091627]
  We present the first approach to use diffusion models as a prior for highly accurate 3D facial BRDF reconstruction from a single image. In contrast to existing methods, we directly acquire the observed texture from the input image, thus, resulting in more faithful and consistent estimation.
  arXiv  Detail & Related papers  (2023-05-10T11:57:49Z)
• DiFaReli++: Diffusion Face Relighting with Consistent Cast Shadows [11.566896201650056]
  We introduce a novel approach to single-view face relighting in the wild, addressing challenges such as global illumination and cast shadows.<n>We propose a single-shot relighting framework that requires just one network pass, given pre-processed data, and even outperforms the teacher model across all metrics.
  arXiv  Detail & Related papers  (2023-04-19T08:03:20Z)
• Shape, Pose, and Appearance from a Single Image via Bootstrapped Radiance Field Inversion [54.151979979158085]
  We introduce a principled end-to-end reconstruction framework for natural images, where accurate ground-truth poses are not available. We leverage an unconditional 3D-aware generator, to which we apply a hybrid inversion scheme where a model produces a first guess of the solution. Our framework can de-render an image in as few as 10 steps, enabling its use in practical scenarios.
  arXiv  Detail & Related papers  (2022-11-21T17:42:42Z)
• A CNN Based Approach for the Point-Light Photometric Stereo Problem [26.958763133729846]
  We propose a CNN-based approach capable of handling realistic assumptions by leveraging recent improvements of deep neural networks for far-field Photometric Stereo. Our approach outperforms the state-of-the-art on the DiLiGenT real world dataset. In order to measure the performance of our approach for near-field point-light source PS data, we introduce LUCES the first real-world 'dataset for near-fieLd point light soUrCe photomEtric Stereo'
  arXiv  Detail & Related papers  (2022-10-10T12:57:12Z)
• Portrait Neural Radiance Fields from a Single Image [68.66958204066721]
  We present a method for estimating Neural Radiance Fields (NeRF) from a single portrait. We propose to pretrain the weights of a multilayer perceptron (MLP), which implicitly models the volumetric density. To improve the generalization to unseen faces, we train the canonical coordinate space approximated by 3D face morphable models. We quantitatively evaluate the method using controlled captures and demonstrate the generalization to real portrait images, showing favorable results against state-of-the-arts.
  arXiv  Detail & Related papers  (2020-12-10T18:59:59Z)
• Deep 3D Capture: Geometry and Reflectance from Sparse Multi-View Images [59.906948203578544]
  We introduce a novel learning-based method to reconstruct the high-quality geometry and complex, spatially-varying BRDF of an arbitrary object. We first estimate per-view depth maps using a deep multi-view stereo network. These depth maps are used to coarsely align the different views. We propose a novel multi-view reflectance estimation network architecture.
  arXiv  Detail & Related papers  (2020-03-27T21:28:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.