Tflite modelmaker object detection: normalizing bounding box values

James_W · November 23, 2021, 5:36pm

I am new to TF lite object detection and have gone through the tutorial:

tensorflow/tensorflow/blob/master/tensorflow/lite/g3doc/tutorials/model_maker_object_detection.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h2q27gKz1H20"
      },
      "source": [
        "##### Copyright 2021 The TensorFlow Authors."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "cellView": "form",
        "id": "TUfAcER1oUS6"
      },
      "outputs": [],
      "source": [

This file has been truncated. show original

and I think I understand it reasonably well. However, I notice that the bounding box is somehow normalised.

So, when I have my own dataset, each row looks like so (only 1 class:

``
TRAIN,/path/to/data//images/0.jpg,ONE_CLASS_CAT,270.5,115.10000000000036,787.3000000000011,483.3000000000011,


How does one normalise these coordinates values for tflite input?

Thank you!