论文标题
极性采样：通过单数值对预训练的生成网络的质量和多样性控制
Polarity Sampling: Quality and Diversity Control of Pre-Trained Generative Networks via Singular Values
论文作者
论文摘要
我们提出了极性采样，这是一种理论上是合理的插入方法，用于控制预训练的深层生成网络DGN的发电质量和多样性）。利用DGN是连续的分段仿射花键可以或可以近似的事实，我们得出了分析DGN输出空间分布，这是DGN的Jacobian单数值的产物的函数，提高到了功率$ρ$。我们将$ \ textbf {polarity} $参数配音，并证明$ρ$将DGN采样集中在模式（$ρ<0 $）或抗模式（$ρ> 0 $）上的DGN输出空间分布。我们证明，非零的极性值比对于许多最先进的DGN的标准方法（例如截断），获得了比标准方法更好的精度重新调整（质量多样性）的帕累托前沿。我们还为改善整体发电质量的改善（例如，就特雷希特的距离而言）提供了定量和定性结果，以实现许多最新的DGN，包括stylegan3，Biggan-Deep，biggan-Deep，NVAE，用于不同的条件和无条件图像生成任务。特别是，极性采样将FFHQ数据集上的stylegan2的最新设备重新定义为FID 2.57，在LSUN CAR数据集中的stylegan2在FID 2.27和afhqv2数据集中的stylegan3至FID 3.95。演示：bit.ly/polarity-samp
We present Polarity Sampling, a theoretically justified plug-and-play method for controlling the generation quality and diversity of pre-trained deep generative networks DGNs). Leveraging the fact that DGNs are, or can be approximated by, continuous piecewise affine splines, we derive the analytical DGN output space distribution as a function of the product of the DGN's Jacobian singular values raised to a power $ρ$. We dub $ρ$ the $\textbf{polarity}$ parameter and prove that $ρ$ focuses the DGN sampling on the modes ($ρ< 0$) or anti-modes ($ρ> 0$) of the DGN output-space distribution. We demonstrate that nonzero polarity values achieve a better precision-recall (quality-diversity) Pareto frontier than standard methods, such as truncation, for a number of state-of-the-art DGNs. We also present quantitative and qualitative results on the improvement of overall generation quality (e.g., in terms of the Frechet Inception Distance) for a number of state-of-the-art DGNs, including StyleGAN3, BigGAN-deep, NVAE, for different conditional and unconditional image generation tasks. In particular, Polarity Sampling redefines the state-of-the-art for StyleGAN2 on the FFHQ Dataset to FID 2.57, StyleGAN2 on the LSUN Car Dataset to FID 2.27 and StyleGAN3 on the AFHQv2 Dataset to FID 3.95. Demo: bit.ly/polarity-samp
