Picture for Michael Soloveitchik

Michael Soloveitchik

Fully Convolutional Fractional Scaling

Add code
Mar 20, 2022
Michael Soloveitchik, Michael Werman

Figure 1 for Fully Convolutional Fractional Scaling
Figure 2 for Fully Convolutional Fractional Scaling
Figure 3 for Fully Convolutional Fractional Scaling
Figure 4 for Fully Convolutional Fractional Scaling

We introduce a fully convolutional fractional scaling component, FCFS. Fully convolutional networks can be applied to any size input and previously did not support non-integer scaling. Our architecture is simple with an efficient single layer implementation. Examples and code implementations of three common scaling methods are published.

Viaarxiv icon
Github IconAccess Paper or Ask Questions
Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon

Conditional Frechet Inception Distance

Add code
Mar 21, 2021
Michael Soloveitchik, Tzvi Diskin, Efrat Morin, Ami Wiesel

Figure 1 for Conditional Frechet Inception Distance
Figure 2 for Conditional Frechet Inception Distance
Figure 3 for Conditional Frechet Inception Distance
Figure 4 for Conditional Frechet Inception Distance

We consider distance functions between conditional distributions functions. We focus on the Wasserstein metric and its Gaussian case known as the Frechet Inception Distance (FID).We develop conditional versions of these metrics, and analyze their relations. Then, we numerically compare the metrics inthe context of performance evaluation of conditional generative models. Our results show that the metrics are similar in classical models which are less susceptible to conditional collapse. But the conditional distances are more informative in modern unsuper-vised, semisupervised and unpaired models where learning the relations between the inputs and outputs is the main challenge.

Viaarxiv icon
Github IconAccess Paper or Ask Questions
Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon