HiFi-Inpaint：面向高保真参考引导图像修复以生成细节保持的人-产品图像
HiFi-Inpaint: Towards High-Fidelity Reference-Based Inpainting for Generating Detail-Preserving Human-Product Images
面向参考图引导的图像修复，提出共享增强注意力（SEA）与基于高频图的细节感知损失（DAL），可直接迁移到需要像素级细节保真的参考驱动编辑/合成任务中。
原文依据 HiFi-Inpaint introduces Shared Enhancement Attention (SEA) to refine fine-grained product features and Detail-Aware Loss (DAL) to enforce precise pixel-level supervision using high-frequency maps.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论展示人与产品融合的人-产品图像在广告、电子商务和数字营销中起着至关重要的作用。生成此类图像的核心挑战在于确保产品细节的高保真保持。在现有范式中，参考引导的图像修复通过利用产品参考图像来引导图像修复过程，提供了一种有针对性的解决方案。然而，现有方法仍存在三个关键方面的局限：缺乏多样的大规模训练数据、当前模型难以聚焦于产品细节的保持、以及粗粒度监督无法实现精确引导。为解决这些问题，我们提出 HiFi-Inpaint，一种专为人-产品图像生成设计的新型高保真参考引导图像修复框架。HiFi-Inpaint 引入共享增强注意力（SEA）以细化细粒度产品特征，并引入细节感知损失（DAL）以利用高频图施加精确的像素级监督。此外，我们构建了一个新数据集 HP-Image-40K，其样本由自合成数据整理并经自动过滤处理。实验结果表明，HiFi-Inpaint 达到了最优性能，能够生成细节保持的人-产品图像。
英文原文摘要 1284 字符
Human-product images, which showcase the integration of humans and products, play a vital role in advertising, e-commerce, and digital marketing. The essential challenge of generating such images lies in ensuring the high-fidelity preservation of product details. Among existing paradigms, reference-based inpainting offers a targeted solution by leveraging product reference images to guide the inpainting process. However, limitations remain in three key aspects: the lack of diverse large-scale training data, the struggle of current models to focus on product detail preservation, and the inability of coarse supervision for achieving precise guidance. To address these issues, we propose HiFi-Inpaint, a novel high-fidelity reference-based inpainting framework tailored for generating human-product images. HiFi-Inpaint introduces Shared Enhancement Attention (SEA) to refine fine-grained product features and Detail-Aware Loss (DAL) to enforce precise pixel-level supervision using high-frequency maps. Additionally, we construct a new dataset, HP-Image-40K, with samples curated from self-synthesis data and processed with automatic filtering. Experimental results show that HiFi-Inpaint achieves state-of-the-art performance, delivering detail-preserving human-product images.