caffe_s3fd icon indicating copy to clipboard operation
caffe_s3fd copied to clipboard

Published 20 hours ago • verified publisher icon lippman1125

how's the inference speed

Open lucasjinreal opened this issue 6 years ago • 3 comments
trafficstars

how about the inference speed of this model?

lucasjinreal avatar Jul 09 '19 11:07 lucasjinreal

vgg16+s3fd hard to be realtime on mobile device, if you want a realtime model, you can you mv2+s3fd. https://github.com/lippman1125/S3FD.PyTorch

lippman1125 avatar Jul 11 '19 01:07 lippman1125

@lippman1125 Not mention mobile, I have test s3fd on GTX 1080TI, it needs about 0.2s per image, which is very slow....

Does this normal?

thanks for your pytorch implementation! seems mobilenet plus s3fd may got reasonable speed! How does the accuracy drop compares to vgg backbone through?

lucasjinreal avatar Jul 11 '19 01:07 lucasjinreal

@lippman1125 Can you share the resnet18 model on Google Drive ?

Thanks.

jamessmith90 avatar Oct 29 '19 14:10 jamessmith90