注意力蒸馏：视觉特征迁移的统一方法
Attention Distillation: A Unified Approach to Visual Characteristics Transfer
AI总结：
本文提出注意力蒸馏损失，利用预训练扩散模型的自注意力特征，在潜在空间中优化合成图像，实现风格、外观和纹理的高效迁移，并集成改进的分类器引导加速生成。
摘要：
近年来生成扩散模型的进展表明，它们在图像风格和语义理解方面具有显著的内在能力。在本文中，我们利用预训练扩散网络的自注意力特征，将参考图像的视觉特征迁移到生成的图像上。与以往将这些特征作为即插即用属性的工作不同，我们提出了一种新颖的注意力蒸馏损失，该损失在理想风格化结果与当前风格化结果之间计算，并基于此在潜在空间中通过反向传播优化合成图像。接着，我们提出了一种改进的分类器引导方法，将注意力蒸馏损失整合到去噪采样过程中，进一步加速合成，并支持广泛的图像生成应用。大量实验证明了我们的方法在将示例的风格、外观和纹理迁移到新图像合成中的卓越性能。代码可在 https://github.com/xugao97/AttentionDistillation 获取。
英文摘要：
Recent advances in generative diffusion models have shown a notable inherent understanding of image style and semantics. In this paper, we leverage the self-attention features from pretrained diffusion networks to transfer the visual characteristics from a reference to generated images. Unlike previous work that uses these features as plug-and-play attributes, we propose a novel attention distillation loss calculated between the ideal and current stylization results, based on which we optimize the synthesized image via backpropagation in latent space. Next, we propose an improved Classifier Guidance that integrates attention distillation loss into the denoising sampling process, further accelerating the synthesis and enabling a broad range of image generation applications. Extensive experiments have demonstrated the extraordinary performance of our approach in transferring the examples' style, appearance, and texture to new images in synthesis. Code is available at https://github.com/xugao97/AttentionDistillation.
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。
