sfd.pytorch
sfd.pytorch copied to clipboard
louis-she
S3FD: single shot face detector in pytorch
hi，I don't seem to see you has done ‘’maxout‘’ in the code. It is different from the paper？
Then I run the command: python3 detector.py --image ./image/test.jpg --model ./epoch_204.pth.tar Then I got this error: Traceback (most recent call last): File "detector.py", line 11, in from model import Net...
In the paper, 3.2 Scale compensation anchor matching strategy stage two is used to compensate the faces with few anchors. However, in anchor.py, mark_anchors seems to compensate all the faces
It cost 100 ms on one GPU by Tesla P100 . How can i do to speed up the infer?
@louis-she HI 在paper中，在计算loss时，有这样的描述： ``` The two terms are normalized by Ncls and Nreg , and weighted by a balancing parameter λ . In our implementation, the cls term is normalized...
@louis-she HI 在利用**anchors坐标**和**网络预测的偏移量**得到**最终box坐标**时，都会引入`一个参数variances`．但在您的代码中，并未看到此参数． ``` x = (predictions[:, 0] * anchors[:, 2] + anchors[:, 0]) y = (predictions[:, 1] * anchors[:, 3] + anchors[:, 1]) w = (torch.exp(predictions[:, 2]) * anchors[:,...
@louis-she HI 结合**caffe源码的[#12](https://github.com/sfzhang15/SFD/issues/12)**和**paper**，感觉： 对于本文的**scale compensation anchor matching strategy**，其是指在降低匹配阈值（0.5到0.35）后，如果image中某个face匹配的anchor数小于6个，才进行2-stage匹配，而且每个face的匹配上限是6． 但在您的代码中，考虑的是**image中所有face**匹配的anchor数（上限100），并不是**单个face**匹配的anchor数
Dear @louis-she, Thank you for your nice work. As **Shifeng Zhang** has mentioned in his paper: `Our anchors are 1:1 aspect ratio (_i.e.,_ square anchor)`, the aspect ratios of all...
