分析用于人脸图像合成的特征提取器网络
Analyzing the Feature Extractor Networks for Face Image Synthesis
AI总结：
本研究针对人脸图像合成评估，分析了 InceptionV3、CLIP、DINOv2 和 ArcFace 等特征提取器在 FID、KID 等指标下的表现，通过多数据集实验揭示其在特征归一化、注意力及域分布上的行为差异。
摘要：
生成对抗网络等技术的进步吸引了研究人员关注人脸图像合成，以生成更加逼真的图像。因此，评估生成图像真实感的标准需求变得显著。虽然配合 InceptionV3 使用的 FID 是基准测试的主要选择之一，但 InceptionV3 在人脸图像上的局限性引发了担忧。本研究考察了多种特征提取器（InceptionV3、CLIP、DINOv2 和 ArcFace）在多种指标（FID、KID、Precision&Recall）下的表现。目标域使用 FFHQ 数据集，源域则使用 CelebA-HQ 数据集及由 StyleGAN2 和 Projected FastGAN 生成的合成数据集。实验对特征进行了深入分析：$L_2$ 归一化、提取时的模型注意力及特征空间中的域分布。我们旨在为评估人脸图像合成方法的特征提取器行为提供有价值的见解。代码公开于 https://github.com/ThEnded32/AnalyzingFeatureExtractors。
英文摘要：
Advancements like Generative Adversarial Networks have attracted the attention of researchers toward face image synthesis to generate ever more realistic images. Thereby, the need for the evaluation criteria to assess the realism of the generated images has become apparent. While FID utilized with InceptionV3 is one of the primary choices for benchmarking, concerns about InceptionV3's limitations for face images have emerged. This study investigates the behavior of diverse feature extractors -- InceptionV3, CLIP, DINOv2, and ArcFace -- considering a variety of metrics -- FID, KID, Precision\&Recall. While the FFHQ dataset is used as the target domain, as the source domains, the CelebA-HQ dataset and the synthetic datasets generated using StyleGAN2 and Projected FastGAN are used. Experiments include deep-down analysis of the features: $L_2$ normalization, model attention during extraction, and domain distributions in the feature space. We aim to give valuable insights into the behavior of feature extractors for evaluating face image synthesis methodologies. The code is publicly available at https://github.com/ThEnded32/AnalyzingFeatureExtractors.
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。
