AIThe Next Webabout 2 hours ago
AI video surveillance learns camera-level normal
1 min read
Lumana's VIA-1 model processes over a billion images daily across 50,000+ cameras, learning what is normal for each camera and flagging deviations. The California startup raised $40 million in July 2025, led by Wing Venture Capital. Most video processing runs locally to cut cloud costs.
Level
Hype check
Tap to vote and see what everyone thinks.
Summary by ByteBrief
