Signature and Log-signature for the Study of Empirical Distributions Generated with GANs
Joaquim de Curtò, Irene de Zarzà, Hong Yan, Carlos T. Calafate
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/decurtoydiaz/signaturesOfficialIn papernone★ 3
Abstract
In this paper, we bring forward the use of the recently developed Signature Transform as a way to measure the similarity between image distributions and provide detailed acquaintance and extensive evaluations. We are the first to pioneer RMSE and MAE Signature, along with log-signature as an alternative to measure GAN convergence, a problem that has been extensively studied. We are also forerunners to introduce analytical measures based on statistics to study the goodness of fit of the GAN sample distribution that are both efficient and effective. Current GAN measures involve lots of computation normally done at the GPU and are very time consuming. In contrast, we diminish the computation time to the order of seconds and computation is done at the CPU achieving the same level of goodness. Lastly, a PCA adaptive t-SNE approach, which is novel in this context, is also proposed for data visualization.
Tasks
Benchmark Results
| Dataset | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| AFHQ Cat | Stylegan2-ada (NVIDIA pre-trained) | MAE Signature | 45,968 | — | Unverified |
| AFHQ Dog | Stylegan2-ada (NVIDIA pre-trained) | MAE Signature | 30,441 | — | Unverified |
| AFHQ Wild | Stylegan2-ada (NVIDIA pre-trained) | MAE Signature | 25,578 | — | Unverified |
| MetFaces | t-Stylegan3-ada (NVIDIA pre-trained) | MAE Signature | 19,872 | — | Unverified |
| MetFaces | r-Stylegan3-ada (NVIDIA pre-trained) | MAE Signature | 22,799 | — | Unverified |
| MetFaces | Stylegan2-ada (NVIDIA pre-trained) | MAE Signature | 23,428 | — | Unverified |
| NASA Perseverance | Stylegan2-ada | MAE Signature | 9,086 | — | Unverified |