search

CHELSEA234 / Multi-domain-learning-FAS

The multi-domain FAS work with SiW-Mv2 dataset (ECCV 2022 oral)
http://cvlab.cse.msu.edu/siwm-v2-dataset.html
68 stars 9 forks source link

How to use recon #12

Closed angelabr closed 1 year ago

angelabr commented 1 year ago

Hi, I don't understand what is the purpose of recon of the liveness image, is it possible or can it be used to obtain kind of a numerical output that states whether an image is live or spoof?

CHELSEA234 commented 1 year ago

@angelabr I am sorry for the late response. We want to obtain the trace by subtracting a spoof image and its corresponding live image. However, we do not have the corresponding live image, so we use recon to micmic the corresponding live image. This imitation is the purpose to have recon.

tangwudu commented 1 year ago

Then how can we use the recon and d_total_loss for live and spoof classification?

tangwudu commented 1 year ago

I think I have found it. Can refer to source_SiW_Mv2/inference.py for final score and decision.

tangwudu commented 1 year ago

But may need some adjustment with the hyper parameters.