0

I'm trying to dig into mediapipe and adapt it to perform inference using a custom tflite model. However, this task seems to be harder than expected. Modifying existing examples is also not practical as they seem to be either too complicated (like pose tracking) or too simple (like hello world).

Can someone provide me with a simple toy example?

Task could be

  • send a float array to the graph (say of length = 1 and all values = 0 for simplicity)
  • tflite model adds 1 to each element of the input tensor
  • output tensor is send back to output and logged to console

Having such a code would surely be a great step ahead while experimenting with the framework

1 Answer 1

0

Meanwhile I solved the issue myself. The solution is also part of MediaPipeUnityPlugin authored by homuler. Here is an example driver code, that shows an example how to use it.

Once I get time I can also post the pure mediapipe solution here (but it's a little ugly)

Sign up to request clarification or add additional context in comments.

Comments

Your Answer

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

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.