PPE Detection Image Annotation Expert

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect skilled annotators with real projects that teach AI systems to see, hear, and understand the world — offering flexible, remote work at the center of a fast-growing industry.

About AI training and this work

AI training (data labeling/annotation) is the human work that helps machine learning models learn from examples. Annotators create the ground-truth data used to train object-detection models — a hands-on way to shape how safety and compliance features behave in real systems.

This project focuses on Personal Protective Equipment (PPE) detection in images — a practical application used in workplace safety, monitoring, and compliance tools.

The role

We’re hiring an experienced data labeling expert to annotate images for a PPE detection model. You will create precise bounding boxes and polygon annotations and assign the correct class labels for each object in the images.

• Employment type: Contractor
• Work location: Remote — worldwide
• Payment: Fixed-price contract — USD 2,000 total
• Data type: Images; Label types: Object detection (bounding boxes) and polygon masks
• Primary tool: Roboflow (experience with CVAT, Label Studio, or similar also useful)

What you’ll do

Your core task is high-quality annotation of images for PPE detection. Accuracy and consistency are essential: every object must be labeled to the agreed classes and annotation quality standards.

• Draw accurate bounding boxes and polygons for each relevant object in an image
• Assign the correct class label to each annotation (see classes below)
• Follow annotation guidelines and quality checks to maintain consistency across the dataset
• Deliver annotations in the required Roboflow-compatible format

Annotation classes and specs

Label each object using the exact class set below. Use polygon masks when precise outlines are necessary; use bounding boxes for general object detection as directed in guidelines.

• Gloves
• Hardhat
• NO-Gloves
• NO-Hardhat
• No-Vest-Overall
• Overall
• Person
• Safety Vest

Requirements

We are seeking an expert-level annotator with proven hands-on experience and a track record of delivering high-quality object-detection datasets.

• Proven experience with image annotation tools such as Roboflow, CVAT, or Label Studio
• Strong attention to detail and ability to maintain high accuracy and consistency
• Experience with object-detection datasets and annotation conventions (bounding boxes, polygons)
• Familiarity with AI datasets for object detection is a plus

How the project works

This is a fixed-price contracting engagement for dataset annotation. You will receive the annotation guidelines, sample images, and access to the annotation workspace (Roboflow). Completed annotations must meet the accuracy and formatting requirements specified in project guidelines.

• Deliverable: annotated image dataset in Roboflow-compatible format
• Label types required: OBJECT_DETECTION and POLYGON
• Client will review annotations for quality and consistency according to provided guidelines

How to apply

If you meet the requirements, apply with a brief summary of relevant experience, the annotation tools you use, and examples or links to past annotation work (if available). OpenTrain connects you to the project and handles application and onboarding.

• Include examples of prior object-detection or PPE-related annotation work if possible
• State your availability and any questions about the annotation process