Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore how AI-powered agents can revolutionize real-time video analysis for open-source intelligence gathering in this conference talk from DEFCON 33's Recon Village. Learn to build and deploy the Autonomous Video Hunter (AVH) system, which combines open-source AI models for image recognition and audio transcription through custom Python-based agents that autonomously detect critical visual elements like logos, scenes, and objects, while simultaneously processing speech keywords and contextual clues from live and recorded video streams. Discover how to cross-reference extracted intelligence with online databases and OSINT repositories to surface actionable leads that would otherwise be lost in overwhelming content volumes, with practical techniques for reducing false positives, handling noisy real-world streams, and scaling analysis across multiple simultaneous video sources. Watch live demonstrations showing AVH identifying target logos and associated context such as social media profiles and geolocation hints from random video clips in seconds, while gaining access to the lightweight open-source AVH tool that you can extend and adapt for your own investigations. Master the design of multi-agent pipelines for real-time video stream analysis, correlation of visual signals and audio cues for OSINT lead generation, efficient scaling across concurrent video sources, and integration of autonomous analysis into existing OSINT workflows through hands-on examples and practical implementation strategies.
