Skip to main content

About the project

Clarity is an AI-powered surveillance system that uses Google’s open-weight Gemma model to detect suspicious activities and potential crimes across multiple security cameras in real time.

Unlike traditional surveillance systems that require constant human monitoring, Clarity intelligently analyzes camera footage to identify potential threats, recognize unusual behavior, and generate alerts when suspicious activity is detected.

One of Clarity’s key features is its ability to track a person’s movement across multiple cameras, reconstructing their path through different locations and creating a visual timeline of where they were observed. By connecting events from different camera feeds, Clarity helps users understand what happened, where it happened, and how an incident unfolded.

Users can also interact with the system through natural language, asking questions such as “What happened near the entrance at 3:45 PM?” or “Where did this person go after leaving the building?” Clarity uses Gemma’s multimodal capabilities to analyze relevant footage and provide timestamped, evidence-based responses.

Our goal is to transform passive security cameras into an intelligent, searchable surveillance network that helps security teams investigate incidents faster, reduce manual monitoring, and improve situational awareness.

Built with

gemmapythongithubapi

AI tools

Coded withCodexClaude CodeGemma 4 26B A4BGemma 4 31B

Prize Categories Entered

  • Deepmind Google

    Best Use of Gemma 4

    Submitted
  • DigitalOcean

    Best Open-Source AI Project

    Submitted