Install YOLOv8 Inside Docker Container in Linux

YOLO (You Only Look Once) is an algorithm for object detection, image segmentation and classification. YOLOv8 is 8th version of YOLO which introduced by Ultralytics in January 2023.

This tutorial explains how to install YOLOv8 inside a Docker container in the Linux. Commands have been tested on Ubuntu.

Prepare environment

Make sure you have installed Docker in your system. If you are using Ubuntu, installation instructions can be found in the post.

Create Docker image

Create Dockerfile and add the following content:

FROM nvidia/cuda:12.0.0-runtime-ubuntu22.04

WORKDIR /yolov8

ADD https://ultralytics.com/assets/Arial.ttf /root/.config/Ultralytics/

RUN apt-get update \
    && apt-get install --no-install-recommends -y  \
        libgl1-mesa-glx libglib2.0-0 python3 python3-pip \
    && pip3 install ultralytics \
    && rm -rf /var/lib/apt/lists/*

ENTRYPOINT ["yolo"]

The nvidia/cuda is used as a base image. So, make sure you have NVIDIA graphics card on the system.

Run command to create Docker image:

docker build -t yolov8 .

Testing YOLOv8

Create directory to store input images:

mkdir -p ~/yolov8/inputs

Download sample image:

curl -o ~/yolov8/inputs/test.jpg https://raw.githubusercontent.com/rdcolema/tensorflow-image-classification/master/cat.jpg

Run the following commands in YOLOv8 container to perform required task:

• Object detection:

docker run -it --rm \
    -v ~/yolov8:/yolov8 \
    yolov8 detect predict save model=yolov8s.pt source=inputs/test.jpg

Output directory: ~/yolov8/runs/detect.

• Image segmentation:

docker run -it --rm \
    -v ~/yolov8:/yolov8 \
    yolov8 segment predict save model=yolov8s-seg.pt source=inputs/test.jpg

Output directory: ~/yolov8/runs/segment.

• Image classification:

docker run -it --rm \
    -v ~/yolov8:/yolov8 \
    yolov8 classify predict save model=yolov8s-cls.pt source=inputs/test.jpg

Output directory: ~/yolov8/runs/classify.

Few results:

Remove Docker image

To remove YOLOv8 image, run the following command:

docker rmi yolov8