Question about detected objects features extraction

WongKinYiu / yolor

implementation of paper - You Only Learn One Representation: Unified Network for Multiple Tasks (https://arxiv.org/abs/2105.04206)

GNU General Public License v3.0

1.99k stars 518 forks source link

Open divid3d opened 2 years ago

divid3d commented 2 years ago

Is it possible to use YOLOR as object detector and feature extractor simultaneously?

WongKinYiu commented 2 years ago

yes.

divid3d commented 2 years ago

Could you please explain a little how to do this? Did you mean feature extraction of each object separately?

chandralegend commented 2 years ago

you can get the output of different layers as the feature vector and then that can be used in your use case.

divid3d commented 2 years ago

I understand, but is it possible to obtain feature vector of every detected object?