Right now, the CNN's negatives training set only includes patches with no overlap with the GT at all. That's a mistake; I should be including more challenging negatives which include the GT, but with a low overlap so that the CNN learns not to activate on off-centre detections (kill some of its spatial invariance!). These aren't "hard negatives" in the traditional sense of "misclassifications produced by a previous version of the detector", but rather just challenging negatives. I think that real hard negatives would be too difficult to implement in the time left.
qxcv
commented
8 years ago
Right now, the CNN's negatives training set only includes patches with no overlap with the GT at all. That's a mistake; I should be including more challenging negatives which include the GT, but with a low overlap so that the CNN learns not to activate on off-centre detections (kill some of its spatial invariance!). These aren't "hard negatives" in the traditional sense of "misclassifications produced by a previous version of the detector", but rather just challenging negatives. I think that real hard negatives would be too difficult to implement in the time left.