Roll Call Vision

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Roll Call Vision

What it does

Volunteer boards approve minutes that record each vote; transcripts often get the count wrong. Roll Call Vision uses OpenCV 5.0 to watch the meeting video during each vote window, count raised hands per seat, and compare with the transcript. An agent then confirms the count, re-looks at the window more densely, or holds minutes approval and asks a human (fail-closed).

Results

Synthetic eval: 32 meetings / 81 votes, 0 unsafe confirms.

Live on AWS

Deployed on AWS Lambda (us-east-2) behind a Function URL: https://hj6wfy2lilbbllikwriblbsx3m0thfif.lambda-url.us-east-2.on.aws/

Note: the Function URL uses AWS_IAM auth, so judges need SigV4-signed requests; the demo video shows the console/live run.

Built with

OpenCV 5, Python, AWS Lambda, Amazon S3, CloudWatch, MCP (agent tools).

Built With

• amazon-web-services
• aws-lambda
• cloudwatch
• mcp
• opencv
• python