Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

I wonder what the latency of the model is. #65

Closed
visionNoob opened this issue Feb 25, 2024 · 1 comment
Closed

I wonder what the latency of the model is. #65

visionNoob opened this issue Feb 25, 2024 · 1 comment

Comments

@visionNoob
Copy link

Thank you for suggesting a good method for real time object-detection.
I have a question
I just looking at the papers in the arXiv but there's no comparison of latency but # params and FLOPs.
Is it because you are still preparing to implement an optimized version (like cuda or something)?
I wonder what the actual latency is like.
Thank you!

@WongKinYiu
Copy link
Owner

#178 (comment)

The main reason we do not show latency is due to different exporting method and even different version of tools will effect too much on latency.

You could take a look yolov6 v3.0 arxiv.

  1. T4 TRT 7: YOLOv7 10% faster than YOLOv6 v3.0
  2. T4 TRT 8: YOLOv7 20% faster than YOLOv6 v3.0
  3. V100 : YOLOv7 more than 30% faster than YOLOv6 v3.0

Many paper show their latency without provide detail information, so we can not verify if the reported latency are fair comparison.

@WongKinYiu WongKinYiu mentioned this issue Mar 7, 2024
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants