Hi there, iam currently working on integration of Bodyposenet(Bodypose2d) model with PoseClassificationNet model and there are few things i need to get addressed.
1-I have utilized Roboflow for annotations of frames for Bodypose2d and then exported them as Coco 17 points Json format. Its working fine and i can utilize that but when it comes to Finetuning PoseClassificationNet we know the output json of BodyPose2d will be input of PoseClassificationNet so how will frames be annotated for PoseClassificationNet independently?
2- Since as the current documentation of PoseClassificationNet say that Output of Bodypose2d will be input of PoseClassifciationNet so if iam annotating frames will i need to add labels too for pose? If yes then how? please guide in detail.