Abstract
We present a comprehensive experimental study on image-level conditioning for diffusion models using cluster assignments. We elucidate how individual components regarding image clustering impact image synthesis across three datasets. By combining recent advancements from image clustering and diffusion models, we show that, given the optimal cluster granularity with respect to image synthesis (visual groups), cluster-conditioning can achieve state-of-the-art FID (i.e. 1.67, 2.17 on CIFAR10 and CIFAR100 respectively), while attaining a strong training sample efficiency. Finally, we propose a novel method to derive an upper cluster bound that reduces the search space of the visual groups using solely feature-based clustering. Unlike existing approaches, we find no significant connection between clustering and cluster-conditional image generation. The code and cluster assignments will be released.
Abstract (translated)
我们提出了一个全面的实验研究，重点关注扩散模型中图像级的处理，使用聚类分配。我们阐明了图像聚类中单个组件如何影响图像合成在三个数据集上的效果。通过结合图像聚类和扩散模型的最新进展，我们证明了，在图像合成（视觉组）的最佳聚类粒度下，聚类条件处理可以实现与当前最佳 FID（即1.67，2.17在CIFAR10和CIFAR100上的值）相同的性能，同时实现强烈的训练样本效率。最后，我们提出了一种新方法，通过仅基于特征的聚类来计算聚类的上界，从而减少了视觉组的搜索空间。与现有方法不同，我们没有发现聚类和聚类条件图像生成之间存在显著的关联。代码和聚类分配将公开发布。
URL
https://arxiv.org/abs/2403.00570
