Closed gs-ren closed 1 year ago
What kind of training scenarios are you supporting with this kind of annotation?
such as pose estimation,face landmarks, which need lines to distinguish each part.
We have considered this operation before, but the operation is not universal enough, and we will develop a feature which can only mark the corresponding list once. This approach will be able to handle such key point ( line | polygon ) tasks
For points annotation, if the connection line between points can be showed, That will be more efficient.Is there such a plan?