Scene Cut Annotation for YouTube Videos (Temporal Segmentation Project)

About OpenTrain

OpenTrain is a central job board for data-labeling and AI-training work. We aggregate openings from many companies and platforms into one place so applicants can discover labeling and human-AI tasks without searching dozens of sites. Creating an OpenTrain account is free and applying takes only a few minutes.

About AI Training and Video Annotation

AI models rely on human-created examples to learn. Video temporal segmentation — marking exact scene boundaries and timestamps — is a common annotation task that helps models understand structure, support summarization, and enable automated editing.

These projects are typically remote and flexible, and they let contributors directly shape how video understanding and editing tools behave.

The Role

We are seeking a qualified data labeling vendor or team of annotators to perform scene cut and transition annotations on a large batch of YouTube videos. You will produce high-quality temporal segmentation labels that mark the exact start and end timestamps of each scene boundary.

• Data type: Video (YouTube)
• Label type: Temporal segmentation (scene boundaries, cuts, transitions)
• Volume: ~1,000 videos (average 5–10 minutes each)
• Project duration: 4–6 weeks
• Preferred team size: 10–15 annotators
• Annotation tool: uLabel (access will be provided) and internal proprietary tooling as instructed
• Expected output: JSON or CSV with fields video_id, scene_id, start_time, end_time
• Employment type: Contractor
• Subject matter: Media & Entertainment (video content editing)

What You'll Do

Follow the supplied annotation guidelines and examples to consistently mark scene boundaries and transitions. Perform internal reviews before delivering finalized files.

• Annotate scene boundaries (cuts and transitions) and mark precise start and end timestamps for each scene.
• Use uLabel (access provided) and any internal tooling as directed by the project team.
• Follow detailed guidelines and example annotations provided by our team.
• Perform internal quality checks and revisions prior to delivery.
• Deliver outputs in the required JSON/CSV format with specified fields.
• Respond to weekly quality audits and implement requested corrections.

Requirements & Proposal Instructions

Submit a proposal that demonstrates your ability to meet the scope and quality expectations. Proposals should be concrete about capacity, process, and pricing.

• Prior experience with video annotation or temporal segmentation tasks is required.
• State your estimated team size and delivery capacity (how many videos per day/week your team can annotate).
• Provide pricing as cost per annotated video or per hour of video annotated.
• Describe your internal QA process and how you will monitor and ensure ≥95% accuracy on spot checks.
• Confirm you can meet agreed turnaround timelines and support weekly quality audits by our internal QA team.

Quality Expectations, Timeline & Pay

Quality standards and punctual delivery are essential. You must pass spot checks and cooperate with weekly audits from our internal QA team.

• Quality requirement: ≥95% accuracy on spot checks.
• Weekly quality audits will be performed by the client's internal QA team; you must correct issues as requested.
• Perform internal QA and revisions before final delivery to the client.
• Project timeline: 4–6 weeks to complete ~1,000 videos (average 5–10 minutes each).
• Preferred staffing: 10–15 annotators to meet throughput needs.
• Pay: USD 4.00 per hour (PAY_PER_HOUR).
• Open to worldwide contractors; hiring and contracting terms will be clarified during selection.