hello~
Thank you for sharing this wonderful work.
Just now, I'm reading your paper and i have a question that how you labeled the datasets ? Or just used the method "via a super-pixel segmentation algorithm." mentioned in the LFW dataset' description?
In addition, could you offer me some suggtions about how to label the datasets?
@Echosanmao LFW dataset has given the corresponding label format, you just need to convert it into the training format (0, 1, 2, ...... N; N is class number).
hello~ Thank you for sharing this wonderful work. Just now, I'm reading your paper and i have a question that how you labeled the datasets ? Or just used the method "via a super-pixel segmentation algorithm." mentioned in the LFW dataset' description? In addition, could you offer me some suggtions about how to label the datasets?