How to include images with no annotation in my dataset - negative samples #4088
Unanswered
tugcekeskin
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hello,
I am trying to do Fracture/Non-fracture bone detection in my project. For Fracture case, there are 4 different classes and they have annotations (both JPG and JSON files). For Non-fracture case, there is no annonation and consists of "JPG" files. I don't know how to include Non-fracture case in my dataset and how to use it with Detectron2. I am using Faster RCNN for object detection.
I've read things about being taken as a negative sample, but I couldn't understand what exactly I should do. How will I create the dataset? I don't want to eliminate the Non-fracture case during training.
Thanks
Beta Was this translation helpful? Give feedback.
All reactions