MirrorPPR：基于样例的人像照片修饰

摘要

虽然基于文本的图像编辑已取得显著进展，但在结构性人像修图方面仍存在局限。文本描述难以精准传达面部特征与身体比例的细微变化。为填补这一空白，我们提出基于示例的人像照片修图任务，模型通过给定示例对，学习推断并将相同修图操作应用于新查询图像。现有基于示例的编辑方法主要聚焦视觉变换显著的任务，而结构性人像修图涉及极为精细的局部修改，使得准确提取与迁移这些编辑操作颇具挑战。为此，我们提出MirrorPPR框架，专用于捕获与迁移细微的结构性修图操作。该方法通过"修图操作提取器"捕捉示例对间的细微差异，并将提取的表示经连接器与低秩适配模块注入预训练的扩散Transformer。此外，跨身份训练对的完美对齐因操作错位而严重受阻。为克服此问题，我们提出先进的数据自增强范式，确保修图操作的严格对齐。为缓解数据稀缺并支持这项新任务，我们构建了包含超过4700万修图对的大规模数据集MirrorPPR47M。通过将数据集划分为模拟子集与专业子集，我们实现了渐进式课程学习以优化网络训练。大量实验表明，MirrorPPR在修图质量与身份保持方面均显著优于现有基线方法。项目页面：https://sjtu-deng-lab.github.io/MirrorPPR。

English

While text-guided image editing has made remarkable progress, it remains limited in structural portrait retouching. Textual descriptions struggle to convey fine-grained changes to facial features and body proportions. To address this gap, we introduce Exemplar-Based Portrait Photo Retouching, where the model is given an exemplar pair and tasked with inferring and applying the same retouching operations to a new query image. Existing exemplar-based editing methods primarily focus on tasks with pronounced visual transformations. In contrast, structural portrait retouching involves extremely delicate and localized modifications, making accurate extraction and transfer of these edits challenging. To tackle this, we propose MirrorPPR, a novel framework designed to capture and transfer subtle structural retouching operations. Our method uses a Retouching Operation Extractor to capture the subtle differences from the exemplar pair. The extracted representations are then injected into a pre-trained Diffusion Transformer (DiT) through a connector and Low-Rank Adaptation (LoRA) modules. Furthermore, constructing perfectly aligned cross-identity training pairs is severely hindered by operation misalignment. To overcome this, we propose an advanced data self-augmentation paradigm that ensures strictly aligned retouching operations. To alleviate data scarcity and support this novel task, we introduce MirrorPPR47M, a large-scale dataset with over 47 million retouched pairs. By structuring the dataset into simulated and professional subsets, we enable progressive curriculum learning to smoothly optimize the network. Extensive experiments demonstrate that MirrorPPR significantly outperforms existing baselines in both retouching quality and identity preservation. The project page is available at https://sjtu-deng-lab.github.io/MirrorPPR.