How to use self annotated object detection boxes for key point recognition in the inference stage.

[SCsleepy]

Hi, I am using rtmpose to identify key points about animals. My question is how to use my annotated object detection box for key point recognition during the inference stage, rather than providing a trained object detection model?

Thank you.

[Ben-Louis]

This function is not yet supported. You could modify the code snippet for bounding box detection in https://github.com/open-mmlab/mmpose/blob/main/demo/topdown_demo_with_mmdet.py#L29-L44 to use annotated box.

[SCsleepy]

Thank you very much for your reply, it has been very helpful to me.