computer vision - How to read the data of Adience benchmark(data set for gender and age classification)? -
i trying train gender , age classification cnn, using data @ adience , got 2 questions.
1 : according website, bounding box of faces recorded in fields "x,y,dx,dy". example, fold_frontal_0_data.txt,the first data is
image name : 10424815813_e94629b1ec_o.jpg (x,y,dx,dy) : 301 105 640 641
however, data of bounding box weird, because size of image 600x601 only, no matter treat (x,y,dx,dy) (left, top, width, height) or (left, top, right, bottom), cannot crop face expected. how crop face properly?
2 : need face alignment on training data? or need face alignment when testing?or both?thanks
Comments
Post a Comment