Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?


Failed to load latest commit information.
Latest commit message
Commit time

Tensorflow Object Detection with Tensorflow 2

Duckies detection

In this repository you can find some examples on how to use the Tensorflow OD API with Tensorflow 2. For more information check out my articles:


You can install the TensorFlow Object Detection API either with Python Package Installer (pip) or Docker, an open-source platform for deploying and managing containerized applications.

First clone the master branch of the Tensorflow Models repository:

git clone

Docker Installation

# From the root of the git repository (inside the models directory)
docker build -f research/object_detection/dockerfiles/tf2/Dockerfile -t od .
docker run -it od

Python Package Installation

cd models/research
# Compile protos.
protoc object_detection/protos/*.proto --python_out=.
# Install TensorFlow Object Detection API.
cp object_detection/packages/tf2/ .
python -m pip install .

Note: The *.proto designating all files does not work protobuf version 3.5 and higher. If you are using version 3.5, you have to go through each file individually. To make this easier, I created a python script that loops through a directory and converts all proto files one at a time.

import os
import sys
args = sys.argv
directory = args[1]
protoc_path = args[2]
for file in os.listdir(directory):
    if file.endswith(".proto"):
        os.system(protoc_path+" "+directory+"/"+file+" --python_out=.")
python <path to directory> <path to protoc file>

To test the installation run:

# Test the installation.
python object_detection/builders/

If everything installed correctly you should see something like:

[       OK ] ModelBuilderTF2Test.test_create_ssd_models_from_config
[ RUN      ] ModelBuilderTF2Test.test_invalid_faster_rcnn_batchnorm_update
[       OK ] ModelBuilderTF2Test.test_invalid_faster_rcnn_batchnorm_update
[ RUN      ] ModelBuilderTF2Test.test_invalid_first_stage_nms_iou_threshold
[       OK ] ModelBuilderTF2Test.test_invalid_first_stage_nms_iou_threshold
[ RUN      ] ModelBuilderTF2Test.test_invalid_model_config_proto
[       OK ] ModelBuilderTF2Test.test_invalid_model_config_proto
[ RUN      ] ModelBuilderTF2Test.test_invalid_second_stage_batch_size
[       OK ] ModelBuilderTF2Test.test_invalid_second_stage_batch_size
[ RUN      ] ModelBuilderTF2Test.test_session
[  SKIPPED ] ModelBuilderTF2Test.test_session
[ RUN      ] ModelBuilderTF2Test.test_unknown_faster_rcnn_feature_extractor
[       OK ] ModelBuilderTF2Test.test_unknown_faster_rcnn_feature_extractor
[ RUN      ] ModelBuilderTF2Test.test_unknown_meta_architecture
[       OK ] ModelBuilderTF2Test.test_unknown_meta_architecture
[ RUN      ] ModelBuilderTF2Test.test_unknown_ssd_feature_extractor
[       OK ] ModelBuilderTF2Test.test_unknown_ssd_feature_extractor
Ran 20 tests in 91.767s

OK (skipped=1)

Running a pre-trained model

The object_detection_tutorial.ipynb notebook walks you through the process of using a pre-trained model to detect objects in an image. To try it out, I recommend to run it inside Google Colab.

Person and Kites detection

Modify code to run on a video stream

The above example can be easily rewritten to work with video streams by replacing the show_inference method with:

import cv2
cap = cv2.VideoCapture(0) # or cap = cv2.VideoCapture("<video-path>")

def run_inference(model, cap):
    while cap.isOpened():
        ret, image_np =
        # Actual detection.
        output_dict = run_inference_for_single_image(model, image_np)
        # Visualization of the results of a detection.
            instance_masks=output_dict.get('detection_masks_reframed', None),
        cv2.imshow('object_detection', cv2.resize(image_np, (800, 600)))
        if cv2.waitKey(25) & 0xFF == ord('q'):

run_inference(detection_model, cap)

Live Object Detection Example

You can find the code as a notebook or python file.

Few-shot learning

The new release also comes with another notebook showing us how to fine-tune a RetinaNet pre-trained model to detect rubber duckies with only 5 images and <5 minutes of training time in Google Colab.

Duckies detection


Gilbert Tanner