Geometry-Aware Single-Image Full-Body Human Relighting

Ji, Chaonan; Yu, Tao; Guo, Kaiwen; Liu, Jingxin; Liu, Yebin

doi:10.1007/978-3-031-19787-1_22

Cited by 7 publications

(2 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some methods have been proposed to enable relighting of human images (Sun et al 2019;Wang et al 2020;Zhou et al 2019;Kanamori and Endo 2018;Pandey et al 2021;Ji et al 2022). However, these image-based methods do not support changing the viewpoints and human poses.…”

Section: Human Relightingmentioning

confidence: 99%

Relightable and Animatable Neural Avatars from Videos

Lin,

Zheng,

Yong

et al. 2024

AAAI

View full text Add to dashboard Cite

Lightweight creation of 3D digital avatars is a highly desirable but challenging task. With only sparse videos of a person under unknown illumination, we propose a method to create relightable and animatable neural avatars, which can be used to synthesize photorealistic images of humans under novel viewpoints, body poses, and lighting. The key challenge here is to disentangle the geometry, material of the clothed body, and lighting, which becomes more difficult due to the complex geometry and shadow changes caused by body motions. To solve this ill-posed problem, we propose novel techniques to better model the geometry and shadow changes. For geometry change modeling, we propose an invertible deformation field, which helps to solve the inverse skinning problem and leads to better geometry quality. To model the spatial and temporal varying shading cues, we propose a pose-aware part-wise light visibility network to estimate light occlusion. Extensive experiments on synthetic and real datasets show that our approach reconstructs high-quality geometry and generates realistic shadows under different body poses. Code and data are available at https://wenbin-lin.github.io/RelightableAvatar-page.

show abstract

Section: Human Relightingmentioning

confidence: 99%

Relightable and Animatable Neural Avatars from Videos

Lin,

Zheng,

Yong

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…A parallel line of work aims to decompose images into geometry and reflectances, enabling physically based relighting [17,18,20,23,39,51]. Recent works leverage the best of learning-based relighting and material decomposition by feeding physics-inspired relighting results from the estimation into another network to produce a final relighting image [19,42,63,68]. Despite plausible relighting results, image-space approaches typically suffer from temporal and view inconsistency artifacts during animation or novel-view rendering, as they lack a 3D parameterization of the scene.…”

Section: Related Workmentioning

confidence: 99%