/Tensorflow-Object-Detection-with-Tensorflow-2.0

Use the Tensorflow Object Detection API with Tensorflow 2

Primary LanguageJupyter NotebookMIT LicenseMIT

Tensorflow Object Detection with Tensorflow 2

This repo was forked from Gilbert Tanner and adds some minor changes to go with my YouTube tutorial series.

How to Install Tensorflow 2 Object Detection:

thumbnail

Tensorflow 2 Object Detection on Webcam and Videos:

thumbnail

Tensorflow 2 Custom Object Detection Model

thumbnail

Count Objects with Tensorflow 2 Object Detection:

thumbnail

Some commands used:

Note: You will probably need to update path names and model names in these commands to match your own environment

Generate TF records

python generate_tfrecord.py --csv_input=images/test_labels.csv --image_dir=images/test --output_path=test.record
python generate_tfrecord.py --csv_input=images/train_labels.csv --image_dir=images/train --output_path=train.record

Start training

python model_main_tf2.py --pipeline_config_path==ssd_efficientdet_d0_512x512_coco17_tpu-8.config --model_dir==training --alsologtostderr

Export graph

python exporter_main_v2.py --trained_checkpoint_dir=training --pipeline_config_path=ssd_efficientdet_d0_512x512_coco17_tpu-8.config --output_directory inference_graph

Potential future work:

  • Extraction of objects detected
  • Recording position data of detections
  • Robot or physical device that uses results of detection (i.e. robot arm)
  • Leave suggestions of other things you'd like to see!

Forked Repo Author

Ben Greenfield

Original Author

Gilbert Tanner