Skip to content
This repository was archived by the owner on Jan 7, 2025. It is now read-only.

device #16

Closed
luhuiwu opened this issue Oct 29, 2020 · 6 comments
Closed

device #16

luhuiwu opened this issue Oct 29, 2020 · 6 comments

Comments

@luhuiwu
Copy link

luhuiwu commented Oct 29, 2020

I have tested it on my computer, but performs is not good. what is wrong?

@zahidzqj
Copy link

me too

@guillaumebrg
Copy link
Contributor

guillaumebrg commented Nov 2, 2020

Hi! Thanks for reporting.

  1. Which script have you tried?
  2. What hardware are you trying to run it on? CPU or GPU specs would be useful.
  3. Can you check your terminal and see if you get any warning messages while running the script?

@zahidzqj
Copy link

zahidzqj commented Nov 5, 2020

I run it on CPU, and I got warning messages *** Unused frames *** when I run 'gesture_recognition.py'. The same result appears in ‘fitness_tracker.py‘, but occasionally you can see the result.

@guillaumebrg
Copy link
Contributor

These *** Unused frames *** messages mean that the inference engine thread is too slow and can't process 16 frames per seconds... as a result, some frames are thrown away. This should explain your performance issues (at least partially). Which script were you testing? Do you maybe have a more powerful CPU you can test it on?

@MoonBunnyZZZ
Copy link

I have tested it on my computer, but performs is not good. what is wrong?

Could you share the downloaded model file?

@MoonBunnyZZZ
Copy link

I run it on CPU, and I got warning messages *** Unused frames *** when I run 'gesture_recognition.py'. The same result appears in ‘fitness_tracker.py‘, but occasionally you can see the result.

Could you share the downloaded model file?

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants