ISee | Devpost

3D Design

Inspiration:

The inspiration behind "ISee: Where AI Meets Vision for the Blind" came from a deep sense of empathy and the desire to make a meaningful impact on the lives of visually impaired individuals. We were inspired by the potential of AI and technology to address real-world challenges, and we wanted to leverage these tools to provide a solution that enhances the independence and safety of blind individuals in their daily lives.

What it does:

"ISee" is a groundbreaking project that utilizes AI and computer vision technology to empower the blind and visually impaired. It integrates a camera and an array of four SONAR sensors to perform real-time object detection, identify and read text from objects, and detect specific signs for navigation purposes. The AI processes this information and conveys it through audio instructions, enabling users to navigate their surroundings more effectively and independently.

How we built it:

We built "ISee" through a combination of hardware and software components. The hardware includes a camera mounted on a hat, SONAR sensors for distance measurement, and audio output capabilities for giving a better idea of the surroundings. On the software side, we employed machine learning models for object detection, text recognition, and sign detection. The entire system is powered by Google Cloud, which provides the necessary computing and storage resources for real-time processing of data. We have used the accelerometer, gyroscope, magnetometer and GPS data from the smart phones to conclude on the walking patterns and wall detection. Only the relevant text boxes are picked up by the AI and easily audible through earphones. We have analyzed the movement patterns and used that user data to distinguish between daily chores and emergency situation. Twilio however is programmed in the system to assist with the emergency situations. While outdoors, the wall navigation protocol is less efficient rather than using GPS data, so for that use case we'll be depending on the GPS co-ordinates and navigation.

Challenges we ran into:

Throughout the development of "ISee," we encountered several challenges. These included optimizing the AI models for real-time performance, ensuring accurate object detection and text recognition, and integrating the hardware components seamlessly. Additionally, managing the data flow and processing in the Google Cloud environment posed its own set of challenges. Overcoming these hurdles required a collaborative and innovative approach.

Accomplishments that we're proud of:

We take pride in creating a solution that has the potential to significantly improve the quality of life for visually impaired individuals. Our accomplishments include developing a robust and user-friendly system that combines cutting-edge technology with real-world accessibility. We're also proud of our commitment to making this technology available and affordable to those who need it most.

What we learned:

Throughout the development of "ISee," we gained valuable insights into the complexities of AI, computer vision, and cloud computing. We learned how to optimize AI models for real-time applications, integrate hardware components effectively, and leverage cloud resources for scalable processing. Additionally, we gained a deeper appreciation for the importance of accessibility in technology.

What's next for ISee:

The journey for "ISee" is just beginning. Our next steps include further refinement and testing of the system to ensure its reliability and accuracy. We aim to collaborate with organizations and communities that support the visually impaired to gather feedback and make necessary improvements. Additionally, we are exploring partnerships to make "ISee" widely available and accessible to those who need it, both locally and globally. Our ultimate goal is to enhance the independence and mobility of visually impaired individuals and continue pushing the boundaries of what technology can achieve in the service of humanity.