About

We help teams build vision AI on their own data

Annotiv is a web platform for collecting consented webcam video from real people, labelling it automatically, and exporting datasets that train modern object-detection models.

Company

Annotiv is operated by [Company legal name], [Registered address]. Contact: hello@annotiv.example.

What we do

• Guided, consented recordings in the browser, with no installs
• On-device auto-labelling with YOLO11 before upload
• A fast review studio for fixing and approving boxes
• Exports for 9 detection models, with training notebooks

Mission

Make good training data easy to get, fairly and safely, so any team can go from idea to a working model in days rather than months.

Values

• Consent first. People know what is recorded and why.
• Least data. Video only, no audio; labelling on-device.
• Your data is yours. Export everything, any time.

What we plan to do

• Team workspaces invite colleagues as reviewers with roles
• More label types polygons/outlines, keypoints and object tracking across frames
• Bring your own model pre-label with a model you trained on Annotiv data
• Hosted training train and evaluate without running a notebook
• Email invites send participant links directly from Annotiv
• Data residency options choose where your data is stored

Roadmap items are plans, not commitments, and may change.