We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
王老师,您好! 根据您相关文件说明 webface和lfw有17个人重叠(/CASIA/webface_lfw_overlap.txt) vgg2和lfw有594个人重叠(/VGGFace2/vggface2_lfw_overlap.txt) 我想请教的是webface和vgg2之间也有重叠吗?这两个数据集是否可以放一起作为训练集? megaface的训练集特别大,470万张图片,672057个人,王老师有没有查找过这个集和LFW的重叠部分? 如果能把webface+vgg2+megaface这三个集放一起作为训练集,那人数还是挺大的。
The text was updated successfully, but these errors were encountered:
其实训练的话就没那么讲究了,你把网络从fc5那分出三路来,分别接三个损失函数更新三个数据集就可以了。
Sorry, something went wrong.
@happynear 这么看insightface用vggface2训练出来lfw结果特别好是不是没有意义啊?
Hi,请问有VGGFace2、MS-Celeb-1M、megaface的数据集吗?能提供个百度云或者谷歌链接吗? @happynear @zhaowwenzhong
No branches or pull requests
王老师,您好!
根据您相关文件说明
webface和lfw有17个人重叠(/CASIA/webface_lfw_overlap.txt)
vgg2和lfw有594个人重叠(/VGGFace2/vggface2_lfw_overlap.txt)
我想请教的是webface和vgg2之间也有重叠吗?这两个数据集是否可以放一起作为训练集?
megaface的训练集特别大,470万张图片,672057个人,王老师有没有查找过这个集和LFW的重叠部分?
如果能把webface+vgg2+megaface这三个集放一起作为训练集,那人数还是挺大的。
The text was updated successfully, but these errors were encountered: