Research video annotation
Review consented study footage using time-coded expression signals that annotators can inspect alongside the recording.
Buyer
Principal investigators, research operations, and behavioral-science labs.
Consented input
Consented interview, study, or interaction video with a documented purpose and capture context.
API role
Process recorded video as asynchronous jobs producing time-coded expression signals for contextual human review.
Human review and action
Researchers define the codebook, inspect clips, resolve uncertain intervals, and retain the participant context.
Business value
Evaluate whether time-coded observations help your annotators locate relevant footage without replacing independent coding.
Proof and validation
Define inter-rater agreement, held-out clips, and an uncertainty policy before evaluating the workflow.
Implementation path
- Define the annotation protocol and consent language
- Submit a representative media sample
- Review a completed job alongside the recording
- Double-code a sample, then review disagreements
Explicit boundaries
- Inferring emotion or thought
- Diagnosing participants
- Automated decisions about people
All workflows Integration readinessBefore you start
Write down the purpose, the people involved, and the question a reviewer will answer. Obtain appropriate consent before recording or processing media. Decide who can access footage and when your team will delete it. These are responsibilities to plan, not a claim that the API supplies a consent or retention system.
Keep the proposed use separate from judging a person's feelings, health, or suitability. For workplace and educational contexts, consider EU AI Act Article 5(1)(f). It prohibits emotion-inference AI there, except for medical or safety reasons, from February 2, 2025. Talk to counsel about your use case. This is not legal advice.
Example review loop
Start with a consented clip and a codebook limited to observable movement. Ask two reviewers to annotate independently before comparing their observations with time-coded model signals. Revisit disagreements in the source footage rather than accepting a model label as the reference. Keep any participant account separate from the movement annotation.
FACS describes anatomically based Action Units, and reliable coding usually relies on certified coders. Agreement between independent coders is recommended. A shortage of manually coded ground truth limits automated Action Unit detection.
A model score is not certified FACS coding. Agreement between annotators does not prove feelings. Keep the observation, the coding decision, and any participant explanation in separate parts of the review record.
How to measure whether it helps
Reserve held-out clips before changing the review protocol. Ask independent reviewers to apply the same written codebook, then inspect disagreements. Noldus The Observer XT supports inter-rater reliability using Cohen's kappa. BORIS is a free, open-source event-logging tool for video, audio, and live observation. These are separate tools, not built-in Mental Edge integrations.
Choose agreement criteria before seeing results, and retain an uncertain or unusable category. Record whether reviewers locate useful moments without mistaking agreement for emotional truth. Report failures alongside useful observations. This is a suggested evaluation loop, not a claim of measured improvement.
What it is not for
Barrett and colleagues describe why expressions should not be read as a person's emotion from the face alone. Keep the explicit boundaries above throughout your proposed implementation. The workflow is not for hiring, promotion, diagnosis, deception detection, threat assessment, student scoring, or surveillance.