Replies: 1 comment 1 reply
-
See https://www.ccoderun.ca/programming/darknet_faq/#negative_samples and https://www.ccoderun.ca/programming/darknet_faq/#image_markup |
Beta Was this translation helpful? Give feedback.
1 reply
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.
-
I have an object detection project of two classes (person, animal), and I want to use darknet/yolov4 for it. The images can contain person(s) and/or animal(s), or none of these classes. For example, an image may just contain cars and I don't need to detect these cars.
My question is how do I handle the annotation file for this image that contains no object of interest? Do I create an annotation file for it and its content is empty?
Beta Was this translation helpful? Give feedback.
All reactions