search

PINTO0309 / OpenVINO-YoloV3

YoloV3/tiny-YoloV3+RaspberryPi3/Ubuntu LaptopPC+NCS/NCS2+USB Camera+Python+OpenVINO
https://qiita.com/PINTO
Apache License 2.0
538 stars 165 forks source link

Syncing the inference to the images #45

Open RachitB11 opened 5 years ago

RachitB11 commented 5 years ago

[Required] Your device (RaspberryPi3, LaptopPC, or other device name):
LaptopPC [Required] Your device's CPU architecture (armv7l, x86_64, or other architecture name):
x86_64 [Required] Your OS (Raspbian, Ubuntu1604, or other os name):
Ubuntu 16.04 [Required] Details of the work you did before the problem occurred:

Followed the instructions

[Required] Error message:

No error

[Required] Overview of problems and questions:
It seems that the inference and the playback rate are almost the same (Around 8.5-9fps) but the inference boxes seems to lag the frame. Is there a way to correctly sync the frames to the inference in imshow?

PINTO0309 commented 5 years ago

Processing asynchronously will shift the frame position slightly, but it will be faster. If you process in synchronization, the frame will not shift, but it will be slightly slower. Use "openvino_yolov3_test.py" or "openvino_tiny-yolov3_test.py" if you want to use synchronous processing.

RachitB11 commented 5 years ago

@PINTO0309 thanks for responding so quickly. I just queried the inputs in the inference request to get the right correlation. On a separate note, how much latency do you observe on running inference per image on the Myriad? Even though the throughput increases with the multiple sticks implementation, the latency for inference will be the same. Do you have any tips on how to speed up the inference?