LensErase：一款用于擦除任意内容的基础模型
LensErase: A Foundation Model for Erase Anything
AI总结：
本文提出基于掩码的通用擦除任意内容基础模型LensErase，构建1.0107亿对示例的数据集，采用Mask-Aware RoPE等技术优化，在多基准上取得显著提升，可作为擦除相关应用的基础模型。
摘要：
我们推出LensErase，这是一款基于掩码的通用“擦除任意内容”基础模型，旨在覆盖广泛的现实世界擦除场景。LensErase支持各类对象、视觉文本、水印、相关视觉效果、多对象移除，以及大范围的对象尺度、宽高比和图像分辨率。首先，为实现如此广泛的覆盖，我们构建了LensErase-100M，这是一个包含1.0107亿对源-掩码-目标示例的大规模数据集，涵盖多样的擦除场景。其次，我们从文本到图像模型Lens预训练LensErase，使其强大的生成先验适配基于掩码的擦除任务。为高效融合源图像与擦除掩码，我们提出了Mask-Aware RoPE，这是一种3D旋转位置编码，可直接将掩码信息注入源图像令牌，无需引入单独的掩码令牌序列。最后，我们通过监督微调与带约束引导、区域感知奖励的强化学习来优化模型，并在推理时应用TraceFusion以保留有效编辑区域外的高保真内容。在六个公开基准及我们提出的LensErase-Bench上，LensErase在基于参考的客观指标与无参考的VLM评估下均取得显著提升。我们希望LensErase能成为擦除相关应用与系统的基础模型。
英文摘要：
We introduce LensErase, a general-purpose mask-based Erase Anything foundation model designed to cover a broad spectrum of real-world erase scenarios. LensErase supports diverse objects, visual text, watermarks, associated visual effects, multiple-object removal, and a wide range of object scales, aspect ratios, and image resolutions. First, to enable such broad coverage, we construct LensErase-100M, a large-scale dataset containing 101.07 million paired source-mask-target examples spanning diverse erase scenarios. Second, we pretrain LensErase from the text-to-image model Lens, adapting its strong generative prior to mask-based erasing. To efficiently incorporate the source image and erase mask, we propose Mask-Aware RoPE, a 3D rotary positional encoding that directly injects mask information into the source-image tokens without introducing a separate mask token sequence. Finally, we refine the model through supervised fine-tuning and reinforcement learning with contract-guided, region-aware rewards, and apply TraceFusion at inference to preserve high-fidelity content outside the effective edit region. Across six public benchmarks and the proposed LensErase-Bench, LensErase achieves substantial improvements under both reference-based objective metrics and reference-free VLM evaluation. We hope LensErase can serve as a foundational model for erase-related applications and systems.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
