So usually in this case for object detection on video data I would load the model on the mobile application and then perform inference on it from there, but I have a scenario where it would be better to have the model loaded on an EC2 instance and using FastAPI to query an endpoint to make predictions on the frames. For example with images I would just send the bytes of the image but for video's I am not sure what to send from the client to the server. Also I realize that it would be better to use a WebSocket over HTTP but just looking for a higher level overview of what to do here.

  • $\begingroup$ Today, how did you solve it? Using the WebSocket solution worked? Or find a better solution? $\endgroup$ Nov 22, 2021 at 18:56


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.