I am working on a project which involves the application of deep learning models. I have collected training data. In collected images, I have more than one object in interest. I am not very clear how to label these images. Should I crop and then label or just feed as it is? I would appreciate any kind of help. In the attached image, I have marked two objects.
1 Answer
$\begingroup$
$\endgroup$
You can create a mask image with the background having value 0 and give all pixels the value of its class.
I am not sure what approach you are taking, but you can for example use Convolutional Neural Networks. They will learn from the mask image what features contribute to a pixel class and can therefore, at inference time, output a per pixel class probability.