StyleGAN has shown remarkable performance in unconditional image generation. However, its high computational cost poses a significant challenge for practical applications. Although recent efforts have been made to compress StyleGAN while preserving its performance, existing compressed models still lag behind the original model, particularly in terms of sample diversity. To overcome this, we propose a novel channel pruning method that leverages varying sensitivities of channels to latent vectors, which is a key factor in sample diversity. Specifically, by assessing channel importance based on their sensitivities to latent vector perturbations, our method enhances the diversity of samples in the compressed model. Since our method solely focuses on the channel pruning stage, it has complementary benefits with prior training schemes without additional training cost. Extensive experiments demonstrate that our method significantly enhances sample diversity across various datasets. Moreover, in terms of FID scores, our method not only surpasses state-of-the-art by a large margin but also achieves comparable scores with only half training iterations.

翻译：StyleGAN在无条件图像生成中展现了卓越的性能。然而，其高计算成本对实际应用构成了重大挑战。尽管近期已有研究尝试在保持性能的前提下压缩StyleGAN，但现有压缩模型仍落后于原始模型，尤其是在样本多样性方面。为解决这一问题，我们提出了一种新颖的通道剪枝方法，该方法利用通道对潜在向量的不同敏感性——这是影响样本多样性的关键因素。具体而言，通过评估通道对潜在向量扰动的敏感性来判断其重要性，我们的方法增强了压缩模型中样本的多样性。由于该方法仅聚焦于通道剪枝阶段，因此能与现有训练方案互补，且无需额外训练成本。大量实验表明，我们的方法在多数据集上显著提升了样本多样性。此外，在FID分数方面，我们的方法不仅大幅超越了现有最优水平，且仅需一半训练迭代即可达到可比分数。