PPR10K: A Large-Scale Portrait Photo Retouching Dataset with Human-Region Mask and Group-Level Consistency

TL;DR

This paper introduces PPR10K, the first large-scale portrait photo retouching dataset emphasizing human-region priority and group-level consistency, along with strategies to improve model performance on these criteria.

Contribution

The paper presents PPR10K, a novel large-scale dataset for portrait retouching with human-region masks and group consistency annotations, and proposes learning strategies to enhance retouching quality.

Findings

Proposed strategies effectively improve HRP and GLC in retouching models.

PPR10K dataset serves as a new benchmark for automatic portrait retouching.

Experiments validate the effectiveness of the learning strategies.

Abstract

Different from general photo retouching tasks, portrait photo retouching (PPR), which aims to enhance the visual quality of a collection of flat-looking portrait photos, has its special and practical requirements such as human-region priority (HRP) and group-level consistency (GLC). HRP requires that more attention should be paid to human regions, while GLC requires that a group of portrait photos should be retouched to a consistent tone. Models trained on existing general photo retouching datasets, however, can hardly meet these requirements of PPR. To facilitate the research on this high-frequency task, we construct a large-scale PPR dataset, namely PPR10K, which is the first of its kind to our best knowledge. PPR10K contains $1, 681$ groups and $11, 161$ high-quality raw portrait photos in total. High-resolution segmentation masks of human regions are provided. Each raw photo is retouched by three experts, while they elaborately adjust each group of photos to have consistent tones. We define a set of objective measures to evaluate the performance of PPR and propose strategies to learn PPR models with good HRP and GLC performance. The constructed PPR10K dataset provides a good benchmark for studying automatic PPR methods, and experiments demonstrate that the proposed learning strategies are effective to improve the retouching performance. Datasets and codes are available: https://github.com/csjliang/PPR10K.