OSOR: 一步扩散修复用于效果感知的对象移除
OSOR: One-Step Diffusion Inpainting for Effect-Aware Object Removal
AI总结：
提出一步扩散模型OSOR，通过占用引导判别器、alpha头和语义锚定验证管道，实现高效、效果感知且对不完美掩码鲁棒的对象移除，速度提升4-30倍。
摘要：
现实世界中的对象移除具有挑战性，原因在于两个关键难点：目标对象的非局部效果（如阴影和反射）难以建模，以及用户提供的掩码通常不准确或不完整。扩散模型拥有数十亿参数和数十个去噪步骤，虽然实现了强大的移除性能，但计算成本高昂，限制了其在交互式应用和边缘设备上的使用。为了解决这些挑战，我们提出了OSOR（一步对象移除），它同时实现了高效、效果感知和掩码鲁棒的对象移除。具体来说，OSOR引入了：(1) 一个占用引导的判别器，用于精确的边界监督，实现稳定的单步扩散训练；(2) 一个alpha头，利用预训练扩散模型的知识，以最小的开销预测适当的移除区域，从而处理不完美的掩码；(3) 一个语义锚定验证管道（SAVP），用于过滤基于噪声指令的三元组，从而大规模生成效果感知的监督。利用SAVP，我们整理了CORNE数据集，包含28万个经过验证的移除对，并进一步标注了AnimeEraseBench和TextEraseBench，以评估在更复杂移除任务上的性能。实验表明，OSOR在感知质量上超越了强大的多步扩散基线，同时实现了4倍到30倍的推理加速。
英文摘要：
Real-world object removal is challenging due to two key difficulties: the target object's non-local effects, such as shadows and reflections, which are difficult to model, and the fact that user-provided masks are often inaccurate or incomplete. With billions of parameters and tens of denoising steps, diffusion-based models achieve strong removal performance at the expense of substantial computational cost, limiting their use in interactive applications and on edge devices. To address these challenges, we present OSOR (One-Step Object Removal), which simultaneously achieves efficient, effect-aware, and mask-robust object removal. Concretely, OSOR introduces: (1) an occupancy-guided discriminator for precise boundary supervision, enabling stable single-step diffusion training; (2) an alpha head that leverages knowledge from pretrained diffusion models to predict appropriate removal regions with minimal overhead, thereby handling imperfect masks; and (3) a semantic-anchored verification pipeline (SAVP) that filters noisy instruction-based triplets to produce effect-aware supervision at scale. Using SAVP, we curate CORNE, which contains 280K verified removal pairs, and further annotate AnimeEraseBench and TextEraseBench to evaluate performance on more complex removal tasks. Experiments show that OSOR surpasses strong multi-step diffusion baselines in perceptual quality while achieving $4\times$ to $30\times$ faster inference.
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。
