EchoGuide

• 

  Logo

Inspiration

The inspiration for this project came from watching a K-drama at the end of 2020, and I remembered that the MC's grandma had developed a condition of receding vision and wanted to help her out. So I thought that using a pre-trained ML model and fine tuning it to detect live objects and then integrating it into a mobile device application could do the trick

What it does

This will be a huge help to the people with receding vision, those who are about to lose their vision, and those who have been blind from birth and cannot be treated medically. The app will alert (through speaker by calling out the name of the object) them about the incoming things (just like in navigation) through motion detection and also the direction in which they must turn to avoid it.

How it can be built

• The first step in this project will be to research and select a suitable object detection model that can run on a mobile device. This will likely involve using a pre-trained model, such as YOLO or SSD, and fine-tuning it to work with the specific types of objects that we want to detect.
• Next, we will develop the mobile application using a framework such as TensorFlow Lite or Core ML. This application will be designed to continuously process video from the device's camera and pronounce the detected objects in real-time (through a text-to-speech API) .
• We will also implement a user interface that allows the user to adjust settings such as the detection threshold, and to hear a history of detected objects.

Challenges we could run into

• The accuracy of the model : For this, we would have to label a large amount of data and train the model using it.

What's next for EchoGuide

1. For visually impaired people: Integrating the model into a mobile application or device (like a smart cane/walking stick) that can provide audio descriptions of the objects in the user's environment.
2. Surveillance and Security: Object detection can be used to automatically identify and track people or vehicles in video footage, which can be useful for monitoring security cameras or detecting suspicious activity.
3. Augmented Reality: Object detection can be used to overlay digital information on real-world objects. For example, an AR app might use object detection to identify a book and display the title and reviews on the screen.
4. Robotics and Autonomous Systems: Object detection can be used to enable robots to navigate and interact with their environment. For example, a robot with object detection capabilities could be programmed to locate and pick up specific objects in a warehouse.

Built With

• ai
• computer-vision
• dl
• ml
• python