Object Detection API

This is a Dockerized Object Detection API that can be used to get the objects boundaries, labels, confidences and face boundaries in an image. As a dockerized app, the api can be installed all OS

---

Example

To send in an image_path
python3 obj_detection_request.py --image="istockphoto-1166584467-612x612.jpg"

To send in an image_url
python3 obj_detection_request.py --image_url="https://c8.alamy.com/comp/HYAMEG/bloemfontein-south-africa-hr-assistant-heather-tookey-enjoying-a-photo-HYAMEG.jpg"

---

Result

{'object_metadata_1': {'label': 'person', 'confidence': 0.83695865, 'object_bbox': [342, 126, 894, 846], 'face_dict': {'face_bbox': (222, 820, 407, 634)}}, 'object_metadata_2': {'label': 'bear', 'confidence': 0.6823649, 'object_bbox': [38, 238, 638, 729]}}

---

To Recreate Result

• Clone the repo
• Change dir to the project folder
• sudo docker-compose build && sudo docker-compose up
• python3 obj_detection_request.py --image="istockphoto-1166584467-612x612.jpg"

---

TO-DO

• Add logging to detect errors