pytorch-deeplab-xception icon indicating copy to clipboard operation
pytorch-deeplab-xception copied to clipboard

Published 20 hours ago • verified publisher icon jfzhang95

inference speed of MobileNet backbone

Open Ruoqianqian opened this issue 5 years ago • 1 comments

It inference 5FPS in 1280 *720 images using mobilenet-deeplab .When I drop the aspp layer, the speed come to 20FPS, Still much slower than the result in mobilenet paper.My GPU is 1080TI.

Ruoqianqian avatar Apr 11 '19 07:04 Ruoqianqian

Maybe the data loading time...

hrtan avatar Oct 23 '19 08:10 hrtan