I'm a noob CV & deep-learning student
I've just started to read a Paper which is about better face detection tech
I'll bring a whole paragraph which I want to ask
we propose Improved Anchor Matching (IAM), which integrates anchor partition strategy and anchor-based data augmentation to better match anchors and ground truth faces, and thus provides better initialization for the regressor.
I roughly understands other sentences but I can't understand initialization for the regressor.
I've googled about that sentence but the only result is for initialization for regression or just regressor.
What does that means? TT
My guess is that regressor is either a fully connected layer or a convolutional layer of the deep learning network layer. So mostly they are talking about the initialization of such layer.
Hope this will of your help.