[Detector Support]: Cat detected as a person #13013
Unanswered
abratchik
asked this question in
Detector Support
Replies: 1 comment
-
Maybe adding own min_score and threshold-values would help you. I think 54% is a real unprecise "detection". I've actually learned that somewhere around 65% is where good detection starts and is not throwing too many false positives for me.
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Describe the problem you are having
Cats are consistently detected as people thus generating false alarms. Example (posting one but I have many more):
![4b8ca7ce3a3c4e0d85524490ecd7e1f8](https://private-user-images.githubusercontent.com/6899165/357336752-5ed87655-f18f-4d1a-9a6a-597702a09b69.jpg?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.E49D0UDSAbcU_xcVt-SyXtAEwL5tsba_LkMvGOSxGgQ)
c51d483f003949e1b7dfe86f0b0d3eb5.mp4
Version
0.14.0-da913d8
Frigate config file
docker-compose file or Docker CLI command
Relevant log output
Operating system
Other Linux
Install method
Docker Compose
Object Detector
RKNN
Any other information that may be helpful
My setup is:
Linux ubuntu 6.1.0-1021-rockchip #21-Ubuntu SMP Mon Jul 29 03:52:32 UTC 2024 aarch64 aGNU/Linux
Orange PI5 Plus rk 3588
Beta Was this translation helpful? Give feedback.
All reactions