Hi there,
I have a simple python script that sends a base64 image to a human detection trained model and return the predictions.
I then add the bounding boxes to the image however what I need to do is convert the image to a png with only the detection’s present.
I suspect there is a way to do this with cv2 however Im wondering if anyone has any more efficient and reliable idea how I can make the image a transparent png showing only the humans my model detected.