HuskyLens is an easy-to-use AI machine vision sensor. It is equipped with multiple functions, such as face recognition, object tracking, object recognition, line tracking, color recognition, and tag(QR code) recognition.
By geab
#43733
I have uploaded 0.4.9Class firmware from the Github and tried to train 3 different objects using a white paper as background.
When I finish the training Huskylens can recognize the 3 objects and differentiate them - Nice!
But whenever I point Huskylens at an area (eg wodden table or the white paper) with none of the 3 objects it will still display eg Object:ID1 or 2 on the screen.

Could you share how you did the training of the gestures in a tutorial?

Cheers, Gernot
By danielmeireles
#44229
geab wrote:I have uploaded 0.4.9Class firmware from the Github and tried to train 3 different objects using a white paper as background.
When I finish the training Huskylens can recognize the 3 objects and differentiate them - Nice!
But whenever I point Huskylens at an area (eg wodden table or the white paper) with none of the 3 objects it will still display eg Object:ID1 or 2 on the screen.

Could you share how you did the training of the gestures in a tutorial?

Cheers, Gernot
I got this behavior after use the object classification, the Huskylens still show Object:ID1 on screen.

Please, this is normal ?. Even if i point to a white paper?.

Thanks