paper-with-me

홈 › Papers

DeRenderNet: Intrinsic Image Decomposition of Urban Scenes with Shape-(In)dependent Shading Rendering

2021-04-28 · Yongjie Zhu, Jiajun Tang, Si Li, Boxin Shi

We propose DeRenderNet, a deep neural network to decompose the albedo and latent lighting, and render shape-(in)dependent shadings, given a single image of an outdoor urban scene, trained in a self-supervised manner. To achieve this goal, we propose to use the albedo maps extracted from scenes in videogames as direct supervision and pre-compute the normal and shadow prior maps based on the depth maps provided as indirect supervision. Compared with state-of-the-art intrinsic image decomposition methods, DeRenderNet produces shadow-free albedo maps with clean details and an accurate prediction of shadows in the shape-independent shading, which is shown to be effective in re-rendering and improving the accuracy of high-level vision tasks for urban scenes.

📄 PDF Abstract BibTeX arXiv:2104.13602

Code (0)

등록된 구현이 없습니다.

Tasks

Intrinsic Image Decomposition

Similar Papers 제목 키워드 기반

Neural Fields meet Explicit Geometric Representation for Inverse Rendering of Urban Scenes

2023-04-06 · Zian Wang, Tianchang Shen, Jun Gao, Shengyu Huang 외

Reconstruction and intrinsic decomposition of scenes from captured imagery would enable many applications such as relighting and virtual object insertion. Recent NeRF based methods achieve impressive fidelity of 3D recon…

3D ReconstructionInverse RenderingNeRF

Neural Fields Meet Explicit Geometric Representations for Inverse Rendering of Urban Scenes

2023-01-01 · CVPR 2023 1 · Zian Wang, Tianchang Shen, Jun Gao, Shengyu Huang 외

Reconstruction and intrinsic decomposition of scenes from captured imagery would enable many applications such as relighting and virtual object insertion. Recent NeRF based methods achieve impressive fidelity of 3D r…

3D ReconstructionInverse RenderingNeRF

Learning Relighting and Intrinsic Decomposition in Neural Radiance Fields

2024-06-16 · Yixiong Yang, Shilin Hu, HaoYu Wu, Ramon Baldrich 외

The task of extracting intrinsic components, such as reflectance and shading, from neural radiance fields is of growing interest. However, current methods largely focus on synthetic scenes and isolated objects, overlooki…

SRUG: Shadow-Guided Relightable Urban Scene with Generation Model

2026-05-23 · Yonghao Zhao, Zexin Yin, Jian Yang, Beibei Wang 외 arxiv

Creating relightable urban scenes from images or videos is widely useful but highly ill-posed. Urban environments are typically unbounded and extend beyond the visible regions. As a result, many portions of the scene rem…

Novel View Synthesis

LightCity: An Urban Dataset for Outdoor Inverse Rendering and Reconstruction under Multi-illumination Conditions

2026-02-01 · Jingjing Wang, Qirui Hu, Chong Bao, Yuke Zhu 외 arxiv

Inverse rendering in urban scenes is pivotal for applications like autonomous driving and digital twins. Yet, it faces significant challenges due to complex illumination conditions, including multi-illumination and indir…

Autonomous DrivingInverse Rendering3D Reconstruction