Observed arrival · 2026-11-20
Tittaeye’s browser-based camera lab
A Vietnamese-language camera app for local image recognition, custom visual labels, alerts, and optional AI features.
- For
- People who want to experiment with browser-based camera monitoring, custom visual labels, and local scene recognition
- Worth noticing
- The page names Florence-2 and CLIP, says local vision needs a roughly 350 MB first download, and notes WebGPU can speed it up.
Field notes
Users can switch between a live camera and a sample video, then add labeled examples by selecting a detection box or drawing a region. The interface exposes separate controls for object, motion, water, fire, smoke, and weather recognition, plus scenario training and alert logs. It says the Florence-2/CLIP vision model downloads locally on first use; optional AI-provider features are separately described as sending images only when the user submits them or enables automatic AI recognition.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue