Trying Object Detection with TFLite Model Maker | imneonizer
In this video, you’ll learn how to train a custom object detection model using TensorFlow Lite Model Maker and deploy it using ONNX runtime.
The thing that makes it interesting is, most of the heavy postprocessing i.e., NMS is already implemented inside the onnx graph it makes the inference very fast.
Colab Notebook: https://colab.research.google.com/drive/1P_ufG0ueE58Xdtm7kaKbSRJrCDOOQdE0?usp=sharing
source
