StyLitGAN: Image-Based Relighting via Latent Control
Anand Bhattad, James Soole, David A. Forsyth
摘要
Generated Image Relit -1 (w + + d 1 ) Relit -2 (w + + d 2 ) Relit -3 (w + + d 3 ) Relit -4 (w + + d 4 ) Relit -5 (w + + d 5 ) Figure 1 . StyLitGAN identifies directional vectors (di) within StyleGAN's style space (W + ) which, when added to the w + style code, effectively modify the lighting of generated images while preserving their geometry and albedo. This process eliminates the need for per-image search or model fine-tuning. The first column displays images generated from StyleGAN2; subsequent columns illustrate the same scene, each relit using a specific direction. These relighting directions (di) are derived through a forward selection method, ensuring diversity and avoiding cherry-picking. The directional effects are consistent across different scenes: for instance, d1 activates an orange-tinged bedside lamp, d2 a less intense white-tinged lamp, d3 introduces strong directional light from the window, and so on, demonstrating diverse relighting capabilities of StyLitGAN.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper14
- UniRelight: Learning Joint Decomposition and Synthesis for Video RelightingKai He, Ruofan Liang, Jacob Munkberg, Jon Hasselgren 等NeurIPS 2025 · 被引用 42 次
- IllumiNeRF: 3D Relighting Without Inverse RenderingXiaoming Zhao, Pratul P. Srinivasan, Dor Verbin, Keunhong Park 等NeurIPS 2024 · 被引用 34 次
- Latent Intrinsics Emerge from Training to RelightXiao Zhang, William Gao, Seemandhar Jain, Michael Maire 等NeurIPS 2024 · 被引用 21 次
- Visual Jenga: Discovering Object Dependencies via Counterfactual InpaintingAnand Bhattad, Konpat Preechakul, Alexei A. EfrosNeurIPS 2025 · 被引用 13 次
- DreamLight: Towards Harmonious and Consistent Image RelightingYong Liu, Wenpeng Xiao, Qianqian Wang, Junlin Chen 等NeurIPS 2025 · 被引用 9 次
它引用的顶会 Paper20
- Alias-Free Generative Adversarial NetworksTero Karras, Miika Aittala, Samuli Laine, Erik Härkönen 等NeurIPS 2021 · 被引用 2,126 次
- Unsupervised Discovery of Interpretable Directions in the GAN Latent SpaceAndrey Voynov, Artem BabenkoICML 2020 · 被引用 459 次
- Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D ScansAinaz Eftekhar, Alexander Sax, Jitendra Malik, Amir ZamirICCV 2021 · 被引用 422 次
- On Aliased Resizing and Surprising Subtleties in GAN EvaluationGaurav Parmar, Richard Zhang, Jun-Yan ZhuCVPR 2022 · 被引用 250 次
- EditGAN: High-Precision Semantic Image EditingHuan Ling, Karsten Kreis, Daiqing Li, Seung Wook Kim 等NeurIPS 2021 · 被引用 248 次
相关 Paper
- StyleGAN knows Normal, Depth, Albedo, and MoreAnand Bhattad, Daniel McKee, Derek Hoiem, David A. ForsythNeurIPS 2023 · 被引用 61 次
- Navigating the GAN Parameter Space for Semantic Image EditingAnton Cherepkov, Andrey Voynov, Artem BabenkoCVPR 2021
- ScribbleLight: Single Image Indoor Relighting with ScribblesJun Myeong Choi, Annie Wang, Pieter Peers, Anand Bhattad 等CVPR 2025
- GANSpace: Discovering Interpretable GAN ControlsErik Härkönen, Aaron Hertzmann, Jaakko Lehtinen, Sylvain ParisNeurIPS 2020 · 被引用 1,049 次
- LumiNet: Latent Intrinsics Meets Diffusion Models for Indoor Scene RelightingXiaoyan Xing, Konrad Groh, Sezer Karaoglu, Theo Gevers 等CVPR 2025
