Tensorflow Instance Segmentation on custom dataset

Hello everyone,

I’m trying to achive instance segmentation on custom dataset which I created in roboflow.
I am using this tutorial: Instance Segmentation with Model Garden  |  TensorFlow Core but I don’t know how to give roboflow data to this system. Is there any example about it?

Hi @Taha_Er ,

Here’s a more accurate explanation of how to use Roboflow data with Instance Segmentation with Model Garden models:

To use annotated data from Roboflow with TFOD models:

  1. Export your annotated dataset from Roboflow in the TFOD format.
  2. Place the exported data in your project directory where the model will run.
  3. Update your model’s configuration to point to this data directory.
  4. Use the TFOD API to load and utilize this data during training or inference.

Hope this Helps .

Thank you !