MoeinSorkhei

Applying the model to map <-> aerial photo

2

Hi, I am interested to apply your model to Google Maps images as demonstrated in your paper. Since I want to evaluate the fake v. real images as you did...

Learned Prior

12

Hi, Thank you for your great implementation. Regarding the "learned" prior, I wanted to ask: 1- Why are you considering the prior to be a Gaussian with **trainable** parameters rather...

Hi, is it possible to provide the JFT-pretrained ResNet-101 checkpoint so we could calculate the domain similarity for other custom datasets?

Hello, The provided links to download pretrained models is now working. Best, Moein

Hi, in your implementation data is always normalised with ImageNet stats: https://github.com/thuml/LogME/blob/main/ranking.py#L75 Since the target dataset might be in a different domain than source (e.g aerial images), the stats of...

Inception-V3 checkpoint not available

1

Hi, thanks for the useful repo. Unfortunately the [link](http://vision.caltech.edu/~macaodha/inat2017/iNat_2017_InceptionV3.pth.tar) provided for the inception pretrained models is not available. Could you provide a valid link? Thanks in advance.

Hello, The meta-data in CMMD dataset does not clearly describe the image paths corresponding to the patient/exam. For instance, patient with D1-0087 has Benign left breast but Malignant right breast...

Hello, thanks for the nice work. RegMean appears to rely on the training data of individual models for computing inner products. Is there a workaround when training data is not...

Questions about the sensitivity function

1

Hello, thanks for providing the code. I have some questions about calculating sensitivity, and I appreciate it if you could clarify them for me. 1. What values of `alpha` and...

Hi, Is it possible to add a ReadMe file so we can understand how to replicate your results and how to apply your model to additional datasets? Thanks in advance.