Why video is the first frontier as AI learns to read the physical world
The Next Web
Read the full article
Lumana, founded by former Intel computer vision leaders, processes over a billion images daily from 50,000+ cameras using its VIA-1 model that learns normal patterns and flags anomalies locally before cloud transmission. With 562 million surveillance cameras outside China by 2025 and two-thirds shipped in 2024 having deep-learning analytics, video is a key area for physical AI leveraging existing infrastructure.
