Object Detection

A FastAPI object detection application based on Yolov5 model.

Getting Started

• Clone the project
• cd into the codebase
• run poetry shell and poetry install to set the virtual environment and install the necessary dependencies

Start the app

Since we used docker, we can start the app by running these commands:

sudo docker-compose build
sudo docker-compose up -d

The app will be served on port 8000

Make request

An example curl request:

curl --location '127.0.0.1:8000/object_detect' \
--form 'input_file=@"/Users/Downloads/download.jpeg"'

Valid response:

{
    "data": [
        [
            {
                "class": 0,
                "class_name": "person",
                "bbox": [
                    134,
                    330,
                    187,
                    579
                ],
                "confidence": 0.7723177671432495
            },
            {
                "class": 0,
                "class_name": "person",
                "bbox": [
                    90,
                    373,
                    140,
                    587
                ],
                "confidence": 0.7619432210922241
            },
            {
                "class": 2,
                "class_name": "car",
                "bbox": [
                    492,
                    314,
                    566,
                    430
                ],
                "confidence": 0.7516090273857117
            },
            {
                "class": 2,
                "class_name": "car",
                "bbox": [
                    379,
                    321,
                    455,
                    460
                ],
                "confidence": 0.7398971915245056
            },
            {
                "class": 0,
                "class_name": "person",
                "bbox": [
                    267,
                    332,
                    315,
                    541
                ],
                "confidence": 0.4890599846839905
            },
            {
                "class": 0,
                "class_name": "person",
                "bbox": [
                    592,
                    322,
                    611,
                    389
                ],
                "confidence": 0.45573562383651733
            },
            {
                "class": 2,
                "class_name": "car",
                "bbox": [
                    444,
                    313,
                    477,
                    400
                ],
                "confidence": 0.38912448287010193
            },
            {
                "class": 0,
                "class_name": "person",
                "bbox": [
                    209,
                    324,
                    260,
                    553
                ],
                "confidence": 0.30747783184051514
            }
        ]
    ],
    "message": "object detected successfully",
    "errors": null,
    "status": 200
}

Error response:

{
    "message": "object detection failed",
    "errors": "error",
    "status": 400
}