I am trying to get an idea of how much data is needed to train a deep convolutional neural network to detect instance masks from images. I am interested in both papers that have been written on the subject and personal experience/advice. Specifically on current state of the art models like Mask RCNN. I know that Mask RCNN was originally trained on MS COCO, but did it need a dataset that large, would a dataset of around 10000-20000 labeled images work?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.