Smartphone screen surface defect annotation

About OpenTrain

OpenTrain aggregates data‑labeling and AI‑training jobs from many platforms into one job board so people can find this kind of work in a single place instead of hunting dozens of sites.

Creating an OpenTrain account is free and applying takes only a few minutes.

About AI training and why it matters

AI training (also called data labeling or annotation) is the human work that teaches models how to understand visual inputs. Labelers prepare examples and correct outputs so models learn to detect real defects and ignore irrelevant background.

This kind of work is commonly remote, flexible, and accessible — many projects need no prior experience and let contributors choose their hours, while giving direct impact on model quality and reliability.

The role

You will create and correct pixel‑accurate segmentation masks that highlight surface defects on smartphone front screens.

Project details from the client: 200 images (each ~20 megapixels) of smartphone front screen surfaces. Defect types include scratches, pit‑defects, blemishes, missing surfaces, cracks, discolored areas, and similar surface damage. Pre‑annotations are provided to speed your work.

• Data type: IMAGE; label type: SEGMENTATION.
• Labeling software: Label Studio.
• Subject matter: smartphone refurbishment/quality inspection.

What you'll do day to day

Work with the provided pre‑annotations and adjust or redraw segmentation masks so they include all defect pixels while excluding background pixels as much as possible.

Follow the project annotation guidelines (shape precision, mask boundaries and defect labeling) inside Label Studio and submit completed annotations per the project workflow.

• Inspect each 20MP image and verify pre‑annotation accuracy before editing.
• Expand or refine masks to fully cover defects (scratches, pits, cracks, discoloration, missing surface, etc.).
• Trim masks to avoid including background or non‑defective regions.
• Save and submit annotations in Label Studio according to project instructions.

Requirements

This is an entry‑level task and there are no other requirements from the client.

The project is open worldwide and accepts remote contributors; you’ll be engaged as a contractor on a part‑time basis.

• Experience level: Entry level — no prior labeling experience required.
• Languages/countries: Worldwide (no location restrictions).
• Employment types: Contractor, Part‑time.
• Time requirement: Less than 20 hours per week.

Compensation and logistics

Pay is hourly: USD $2.00 per hour (PAY_PER_HOUR). Hours and scheduling are flexible within the less‑than‑20‑hours/week limit.

You will use Label Studio to view images, edit segmentation masks, and submit work. Pre‑annotations are provided to reduce initial annotation time.

• Hourly rate: $2.00 USD per hour.
• Project size: 200 images (~20MP each).
• Labeling tool: Label Studio (access provided by the project).

How to apply

Create a free OpenTrain account and submit your application for this role — applying takes only a few minutes.

Because this project accepts entry‑level contributors worldwide, applications will be reviewed based on availability and fit with the project workflow.

• Sign up at OpenTrain and apply through the listing page.
• If selected, you will receive project access and labeling instructions inside Label Studio.