Has anyone tested the time-consuming of the code?

ChenYingpeng / caffe-yolov3

A real-time object detection framework of Yolov3/v4 based on caffe

473 stars 231 forks source link

Has anyone tested the time-consuming of the code? #17

Open mayidu opened 5 years ago

mayidu commented 5 years ago

hello, I see the code process an image use time is (52+252)ms, which 52ms is processing data, and 252ms is network. Has anyone tested the time-consuming of the code?I want to know if my test is correct.

ChenYingpeng commented 5 years ago

This key is your device! Averaging inference 10 times is (30 + 59) ms on GTX1060.

evercherish commented 5 years ago

Hello. My inference time is (110+150)ms on GTX1080Ti while detecting a 608x608 image. Is there anything wrong when I build the projetc? Thanks!

deeplearning666 commented 4 years ago

Hello. My inference time of detection is 155ms on GTX1060 while detecting a 608x608 image on yolov4. However, when I use darknet, the inference time is 80ms. Is there anything wrong when I build the projetc? Thanks! I'm looking forward to hearing from you soon!