HumanOLAT: A Large-Scale Dataset for Full-Body Human Relighting and Novel-View Synthesis

TL;DR

The paper introduces HumanOLAT, a comprehensive large-scale dataset of multi-view full-body human captures with varied lighting, aiming to advance research in human relighting and novel-view synthesis.

Contribution

It provides the first publicly available high-quality OLAT dataset for full-body humans, enabling benchmarking and development of relighting and rendering methods.

Findings

Evaluations highlight the dataset's usefulness for current methods.

Challenges remain in modeling complex human lighting interactions.

The dataset facilitates future research and benchmarking.

Abstract

Simultaneous relighting and novel-view rendering of digital human representations is an important yet challenging task with numerous applications. Progress in this area has been significantly limited due to the lack of publicly available, high-quality datasets, especially for full-body human captures. To address this critical gap, we introduce the HumanOLAT dataset, the first publicly accessible large-scale dataset of multi-view One-Light-at-a-Time (OLAT) captures of full-body humans. The dataset includes HDR RGB frames under various illuminations, such as white light, environment maps, color gradients and fine-grained OLAT illuminations. Our evaluations of state-of-the-art relighting and novel-view synthesis methods underscore both the dataset's value and the significant challenges still present in modeling complex human-centric appearance and lighting interactions. We believe HumanOLAT will significantly facilitate future research, enabling rigorous benchmarking and advancements in both general and human-specific relighting and rendering techniques.