Edit2Perceive：图像编辑扩散模型是强大的密集感知器
Edit2Perceive: Image Editing Diffusion Models Are Strong Dense Perceivers
把图像编辑扩散模型（FLUX.1 Kontext）当作密集预测骨干，用像素空间一致性损失约束中间去噪状态、单步确定性推理出深度/法线/抠图，为视觉研究者提供一条替代 T2I 生成式骨干的 image-to-image 密集感知路线。
原文依据 image editing diffusion models are inherently image-to-image consistent, providing a more suitable foundation for dense perception task
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期扩散 Transformer 的进展在视觉合成上展现出显著的泛化能力，然而大多数密集感知方法仍依赖为随机生成设计的文本到图像（T2I）生成器。我们重新审视这一范式，并表明图像编辑扩散模型本质上是图像到图像一致的，为密集感知任务提供了更合适的基础。我们提出 Edit2Perceive，一个统一的扩散框架，将编辑模型适配到深度、法线与抠图任务。基于 FLUX.1 Kontext 架构，我们的方法采用全参数微调与像素空间一致性损失，以在中间去噪状态之间实现保持结构的细化。此外，我们的单步确定性推理在相对较小的数据集上训练的同时，可将运行时间最高提升至更快。大量实验表明，在全部三项任务上均取得全面的最优结果，揭示了面向编辑的扩散 Transformer 在几何感知方面的强大潜力。
英文原文摘要 1039 字符
Recent advances in diffusion transformers have shown remarkable generalization in visual synthesis, yet most dense perception methods still rely on text-to-image (T2I) generators designed for stochastic generation. We revisit this paradigm and show that image editing diffusion models are inherently image-to-image consistent, providing a more suitable foundation for dense perception task. We introduce Edit2Perceive, a unified diffusion framework that adapts editing models for depth, normal, and matting. Built upon the FLUX.1 Kontext architecture, our approach employs full-parameter fine-tuning and a pixel-space consistency loss to enforce structure-preserving refinement across intermediate denoising states. Moreover, our single-step deterministic inference yields up to faster runtime while training on relatively small datasets.Extensive experiments demonstrate comprehensive state-of-the-art results across all three tasks, revealing the strong potential of editing-oriented diffusion transformers for geometry-aware perception.