The two papers have inconsistencies in their explanations of DC
Opened this issue · 0 comments
btt-creator commented
Hi，author！thanks your contribution on the PPFP! when i was reading “Privacy-Preserving Face Recognition with Learnable Privacy Budgets in Frequency Domain”， it mentioned that when we remove the DC component, it is impossible to reconstruct the image without knowing the DC component.But in this paper "Privacy-Preserving Face Recognition Using Random Frequency Components",the article states that it is possible to reconstruct the original image using only high-frequency information.Could you please explain how to reconstruct using only high-frequency information?