The running of openvino+HDDL-R8 is slower than i3

TNTWEN / OpenVINO-YOLOV4

This is implementation of YOLOv4,YOLOv4-relu,YOLOv4-tiny,YOLOv4-tiny-3l,Scaled-YOLOv4 and INT8 Quantization in OpenVINO2021.3

MIT License

238 stars 66 forks source link

The running of openvino+HDDL-R8 is slower than i3 #6

Closed lijunnankman closed 4 years ago

lijunnankman commented 4 years ago

I am Kman.When i use the HDDL-R8 accelerator card to run the official demo.The fps are very low,and it showed a very low "util%". My hardware configuration is as follows:

CPU: i3-8145UE Vision Accelerator Card: hddl-R8 OS: Ubuntu18.04 OpenVINO:2020.4

Have you met this situation?

TNTWEN commented 4 years ago

Hello! You could see my detailed test of AI-BoxX Gen.1(https://github.com/TNTWEN/OpenVINO-YOLOV4/issues/5) GPU is faster than HDDL-R8.AndI if you need to run other programs at the same time, I don't recommend that you use the CPU to do inference. Thanks for your trying！

TNTWEN commented 4 years ago

In this project,GPU - Intel UHD Graphics 620 (WHL GT2)>HDDL-R8>CPU - Intel Core i3-8145UE CPU @ 2.20GHz × 4 if you run YOLOv4 correctly. Until now ,i haven't test old YOLOV3 using intel AI-BoxX Gen.1.

lijunnankman commented 4 years ago

@TNTWEN Thank you for your reply. I agree with your recommention.However, when I run the intel demo(face_recognition_demo) in HDDL-R8,the fps is slower than i3. When I check the HDDL-r8's util% ,they are all less than 10%. Btw, my intel AI-BoxX Gen.1. is from Uzel.