Is it possible to merge two Yolov7 models to inference in just one instance?

Question

I have two Yolo v7 models (*.pt weights) that I need to merge:

• Pose estimation https://github.com/RizwanMunawar/yolov7-pose-estimation
• Model that detect helmets

Is it possible to create a model from the weights of the models I am talking about?

I need the same video stream to detect the pose of the person and at the same time to detect if he/she is wearing a helmet or not.

I tried to run together but I have some memory issues, that's why I need just con *.pt file

Thank you.