You are really swimming against the current by using Node (or Java, or other languages) and not Python or C++ to do things with OpenCV. There are so many resour

You are really swimming against the current by using Node (or Java, or other languages) and not Python or C++ to do things with OpenCV. There are so many resources that you have to painfully translate to Javascript. Plus, almost every other thing you would want to hook up to a pipeline containing face detection (like face recognition, tensor flow object detection, etc) is easily accessed from Python.