image annotations flip?

Open Cv9527 opened this issue 8 years ago • 3 comments

@Seanlinx 152 line in imdb.py: m_bbox[0], m_bbox[2] = -m_bbox[2], -m_bbox[0]. It seems the box is fliped. does the flip is necessary when training model?

Apr 11 '17 09:04 Cv9527

@Cv9527 Not necessary, but adding mirrored images will enlarge the training set and improve the performance.

Apr 11 '17 09:04 Seanlinx

during training Pnet, for the face class task, how to ignore part label -1 when calculate softmaxloss, for regression task, how to ignore negative data?

Apr 11 '17 10:04 Cv9527

@Cv9527 The negativemining layer sets the out_grad to "0" where it should be ignored.

Apr 11 '17 11:04 Seanlinx