All videos
0:00 / 0:00
ai

AI Dev 26 x SF | Thierry Damiba: Edge to Cloud Video Anomaly Detection

DeepLearning.AI16 June 2026Watch on YouTube

Description

This talk by Qdrant's Thierry Damiba shows how to build a real-time video anomaly detection system that works in open-world settings, where the most important events are often the ones you did not explicitly train for. It walks through an edge-to-cloud architecture using Qdrant Edge, Twelve Labs, and NVIDIA Metropolis to detect unusual behavior, search video semantically, and support grounded investigation workflows.

What you'll learn

  • Learn how to build a real-time video anomaly detection system with edge-to-cloud architecture for detecting unusual behavior
  • Understand how Qdrant Edge, Twelve Labs, and NVIDIA Metropolis work together in an integrated video analysis pipeline
  • Discover how the system detects unexpected events it wasn't explicitly trained for in open-world environments
  • See how semantic video search and grounded investigation workflows support investigative work

Frequently asked questions

What are the benefits of an edge-to-cloud architecture for video anomaly detection?
Edge-to-cloud architecture enables real-time processing close to the source while allowing the cloud to handle more advanced analysis and storage, making it more effective for detecting unusual behavior in video surveillance.
How does this system help you find unexpected events you didn't anticipate?
The system is designed to work in open-world environments where it can detect unusual behavior without being explicitly trained for it, which is important because the most critical events are often the ones you didn't foresee.
What role does semantic video search play in the investigation process?
Semantic video search allows investigators to search videos based on meaning and content rather than just keywords, which supports grounded investigation workflows.
What technologies form the core of this video anomaly detection system?
The system uses Qdrant Edge for vector search, Twelve Labs for video understanding, and NVIDIA Metropolis for computer vision processing, which together enable real-time anomaly detection and analysis.

Topics

anomaly detectioncomputer visionedge computingvideo analysisai applications

In this video

Nvidia

Related reads