Inspiration
Our project is motivated by the fact that people go missing daily, often due to human trafficking and other harmful activities. This results in families being separated, and the global crisis has made these problems worse, causing trafficking to become more diverse and widespread. Our goal is to bring families back together and address these serious issues.
What it does
Our solution utilizes video security feeds to locate missing persons in crowded areas such as hotels and conventions, where keeping track of everyone can be difficult. By leveraging existing infrastructure effectively, it enables authorities and security personnel to respond promptly and efficiently to such situations. Our approach leans heavily on a chat-style interaction, departing from the traditional investigative process. Instead of relying solely on formal procedures, our tool facilitates communication in a conversational manner. It fosters direct exchanges between security personnel and the system, allowing for real-time updates and immediate responses. This informal approach streamlines the process, making it more intuitive and user-friendly, thus enhancing efficiency and effectiveness in locating missing persons.
How we built it
By harnessing the power of Gemini Pro 1.5 models from Vertex AI, along with integrating Google Maps API and collecting vital data such as device information, floorplan details, and device-floorplan mapping, our Vertex AI agent successfully achieved the project's objective. This comprehensive approach to utilizing various resources allowed us to effectively process video data, enabling the agent to fulfill its designated tasks.
Challenges we ran into
Firstly, we had to narrow down where our solution could be most effective, like in busy places such as airports or malls. Secondly, we had to ensure we protected people's privacy while using video data, so we implemented strong anonymization techniques and restricted access. Lastly, staying up-to-date with the ever-changing regulations about using this kind of technology in different locations was crucial to our success.
Accomplishments that we're proud of
We're tackling a problem we know is important because we've seen the stats that show it's a big issue. We've brainstormed some great ideas to solve it, drawing from real-world problems to make sure our solution is practical and effective.
What we learned
From this project, we learned some really important stuff! First off, we realized how vital it is to do our research thoroughly, especially by looking at stats and real-life situations. Working together as a team taught us that brainstorming and sharing ideas can lead to some awesome solutions. Dealing with sensitive info, like videos, showed us that we've got to be super careful about people's privacy and put strong protections in place. Also, we learned how regulations can change, so we need to stay flexible and adapt. Overall, this project gave us a ton of valuable lessons that we can use in the future.
What's next for VertexGuard
In the pipeline for this project is the expansion of its reach to more locations, potentially teaming up with other organizations to bolster its capabilities further. Basically, we're going to keep improving and growing so we can help more people and keep places safe!
Built With
- gcp
- gemini-pro-1.5
- google-maps
- vertex
Log in or sign up for Devpost to join the conversation.