Get results as YOLO annotation about ultralytics

[github-actions]

👋 Hello @gillonba, thank you for your interest in Ultralytics YOLOv8 🚀! We recommend a visit to the Docs for new users where you can find many Python and CLI usage examples and where many of the most common questions may already be answered.

If this is a 🐛 Bug Report, please provide a minimum reproducible example to help us debug it.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset image examples and training logs, and verify you are following our Tips for Best Training Results.

Join the vibrant Ultralytics Discord 🎧 community for real-time conversations and collaborations. This platform offers a perfect space to inquire, showcase your work, and connect with fellow Ultralytics users.

Install

Pip install the ultralytics package including all requirements in a Python>=3.8 environment with PyTorch>=1.8.

pip install ultralytics

Environments

YOLOv8 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

• Notebooks with free GPU:
• Google Cloud Deep Learning VM. See GCP Quickstart Guide
• Amazon Deep Learning AMI. See AWS Quickstart Guide
• Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all Ultralytics CI tests are currently passing. CI tests verify correct operation of all YOLOv8 Modes and Tasks on macOS, Windows, and Ubuntu every 24 hours and on every commit.

[glenn-jocher]

Hello! It sounds like you're looking to automate the process of saving detections with low confidence scores for further annotation. Currently, YOLOv8 does not directly output detections in YOLO annotation format, but you can easily extract this information from the Results object and format it accordingly.

Here's a quick example of how you might do this in Python:

from ultralytics import YOLO

# Load your model
model = YOLO('path/to/your/custom_model.pt')

# Run predictions
results = model.predict(source='path/to/your/images')

# Save low confidence detections
threshold = 0.25  # Set your confidence threshold
for result in results:
    for box in result.boxes:
        if box.conf < threshold:
            with open('low_conf_detections.txt', 'a') as file:
                # Format: class x_center y_center width height
                x_center, y_center, width, height = (box.xywh / result.orig_img.shape[1::-1]).tolist()
                file.write(f'{int(box.cls)} {x_center} {y_center} {width} {height}\n')

This script will append the low confidence detections to a text file in the YOLO format. You can adjust the threshold as needed to capture the detections you're interested in reviewing. Hope this helps!