Democratise public agency user access to AI/ML on VAS
summary
GovTech builds VAS on AWS to put ML next to video streams, claiming it handles use cases from post‑mortem trend reports to light realtime image analytics and heavier real‑time inference. Platform uses Amazon Rekognition, SageMaker/SageMaker Pipelines and SageMaker Canvas, offers a no‑code VideoIO for dataset curation and training, a reporting Insights module, and a forthcoming Cloud Video Exchange. It cites 200M IP cameras, ~1,000 PB/day growing 35% and only 2% of video analyzed; ~10 agencies already trial models for crowd counting, dementia screening, larvae and hatchery analysis.
AI-written summary · about
chapters (19)
- 0:00Introduction and speakers
- 1:24Problem statement on video analytics
- 3:12AWS ML stack and SageMaker
- 5:36Amazon Rekognition and VAS
- 7:05GovTech’s video analytics journey
- 10:31Core areas for video analytics
- 12:41Adoption challenges in public sector
- 14:08GovTech’s AI team and approaches
- 17:07Agencies’ maturity in video analytics
- 19:37Operationalizing video analytics
- 22:04Use case examples from GovTech
- 25:58What is VAS platform
- 28:48VAS personas and components
- 30:36VAS value propositions
- 32:22VAS tech stack and ecosystem
- 34:31Next steps for agencies
- 36:21AWS training resources
- 37:24Key takeaways summary
- 37:45Q&A with agencies onboarding
related talks
28mVision, Big Data and AI
30mAI in the Cloud
17mAI/ML demo derby
26mUse computer vision at the edge to improve operations with AWS Panorama
36mSolve common business problems with AI/ML
46mComputer Vision to Secure Your Surroundings with AI/ML Solution Built Using Open Source Tools at the Edge
13mPremiere
29mSolve common business problems with AI and ML