Xuan Lin
Xuan Lin
"-1" is the default value of label that would be ignored in SoftmaxOutput layer if use_ignore=True. And landmark face should be ignored in classification task. @so-as
@so-as Then how could you distinguish them from part faces
@so-as That's not the point. I mean you should provide two kinds of label, one is 'raw_label' that contains '-2', the other is 'cls_label' which only contains '0', '1' and...
@Cv9527 No. raw_label = 1, 0, -1, -2, which is used to distinguish cls, bbox and landmark in negativemining layer. Since softmax_output layer only accpet label value in [1, 0,...
@Cv9527 Right, but 'label' is not needed in negativemining layer since you've got 'raw_label'.
@Cv9527 `return in_shape, [in_shape[0], in_shape[1], keep_shape, keep_shape, keep_shape]`
@Cv9527 Sorry, it should be `return in_shape, [in_shape[0], in_shape[1], in_shape[2], keep_shape, keep_shape, keep_shape]`
@Cv9527 Yes, and it should be almost the same as the one for bbox loss
The fuction 'gt_imdb()' will return negative dataset as long as the imglist file in .data/mtcnn/imglists/ contains negative images
Negative image's gt_bbox is set to 0,0,0,0, so the gradient can be computed normally. And I block the gradient of negative images in the customOp named 'negativemining'.