Face Detection + Gender Classification: Pipelining two models on Hailo devices

Many thanks!
I will try this out and report back to you (although this will probably take some time because I am currently focusing on streamlining my training pipeline for my detection model) :slight_smile:

Hi @Simon_Keilbach
Just curious: the accuracy of the face detection models already in our zoo not good enough for your use case?

Hi @shashi
I have sent you a dm regarding your question

Hello @shashi I’m facing Inaccuracy issues while detecting age and gender using the suggested models.

For gender: Sometimes it detects men faces as women, even if the person is in front of the sensor and even if the person have beard and mustache.

For age: most of the times it calculates people between 30 and 40 years, even if they older almost between 50 and 60 years.

Is there a way to increase accuracy?

@shashi this is an example following this tutorial I’m detected as female


another example