Frechet Inception Distance
FID
作画AIの性能指標
数値が低いほうが性能が高い
いろいろな欠点があるらしいhttps://www.businessinsider.jp/post-257703
2022 Weights & Biases
Supervised learningは評価が簡単
予測と出力を比較すればいい
GANで生成された画像はどう定量化すればいいのか？
2つの軸を考えてこれらの評価法を考えていく
Fidelity: We want our GAN to generate high quality images.
Diversity: Our GAN should generate images that are inherent in the training dataset.
よくあるアプローチ
Pixel Distance:
This is a naive distance measure where we subtract two images' pixel values.
However, this is not a reliable metric.
Feature Distance
We use a pre-trained image classification model and use the activation of an intermediate layer.
This vector is the high-level representation of the image. Computing a distance metric with such representation gives a stable and reliable metric.
FIDではこっちを使う