http://bcsiriuschen.github.io/CARC/
不包含1：
The dataset metadata and features (4.4G)
Original face images (detected and croped by openCV face detector)
16 faical landmark locations
包含Verification Subset (CACD-VS)
Image pairs in verification subset (CACD-VS) (198M)
High dimensional LBP features (in .mat format) for CACD-VS (192M)
Script (gnuplot) and data to generate the ROC curves in our TMM paper (Fig. 9) (18K)
