Visionaire-Ai

See world through Visionaire-Ai.

Comment

Inspiration

The inspiration behind Visionaire-AI is the daily challenges faced by visually impaired people when identifying objects and moving around unfamiliar environments independently. You noticed that simple everyday tasks—such as finding a bottle, locating belongings, and judging how far away an object is—can require additional assistance for people with visual impairments. This inspired you to develop an AI-powered assistant that uses a smartphone camera, voice commands, and object detection to help users identify objects.

How it work

In Visionaire-AI, the system can guide the user about whether an object is on the left, right, or centre of the camera view. Here is how it works in simple language. How left, right, and centre detection works

  1. Object on the left The AI detects the object on the left side of the camera screen. Voice feedback: “The bottle is on your left.”

  2. Object in the centre The AI detects the object near the middle of the screen. Voice feedback: “The bottle is in front of you.”

  3. Object on the right The AI detects the object on the right side of the camera screen. Voice feedback: “The bottle is on your right.”

How does the technology determine the direction? Mobile camera captures the scene

YOLOv8 detects the object It draws a bounding box around the object.

JavaScript checks the box position It compares the box's horizontal centre with the centre of the screen.

Voice assistant speaks the direction

What it does

Visionaire-AI is a smart assistant that helps visually impaired people identify objects and understand how far away they are. It uses a mobile camera to detect objects, such as a bottle, and speaks instructions aloud. An ESP32 with a VL53L1X distance sensor helps measure the distance to an object.

How we built it

We built Visionaire-AI by combining software, artificial intelligence, and hardware. How we built it

  1. HTML and CSS: We created the website interface, including the camera view, information box, and distance display.
  2. JavaScript: We programmed voice commands so users can say things like “Open camera” and “Find bottle.”
  3. YOLO AI model: We used YOLOv8 to detect objects through the mobile camera and display a box around the detected object.
  4. ESP32: We used an ESP32-S3 microcontroller to connect the distance sensor to the system through Wi-Fi.
  5. VL53L1X sensor: This sensor measures the distance between the sensor and an object.
  6. Voice feedback: We used speech technology to speak object names and provide alignment instructions.
  7. Integration: We connected the camera-based detection system with the hardware-based distance measurement to create one assistive system.

Challenges we ran into

AI Object Detection: Making the AI detect objects correctly using the mobile camera.

Voice Commands: Making the system understand voice commands and respond with spoken instructions.

Distance Measurement: Connecting the VL53L1X sensor to the ESP32-S3 to measure the distance of objects accurately.

Hardware and Software Integration: Connecting the camera-based AI system with the ESP32 and distance sensor so they could work together.

Wi-Fi Connectivity: Sending distance data from the ESP32 to the website and handling connection issues.

Debugging and Testing: Fixing errors in the code, Arduino IDE, and communication between different components.

Accomplishments that we're proud of

Social Impact: Designed the project to support visually impaired people in identifying everyday objects and becoming more independent.

Competition Recognition: Achieved a place in the Top 200 of Vivo Ignite 2026 among submissions from across India.

What's next for Visionaire-Ai

In the future, we plan to improve Visionaire-AI by making object detection more accurate, improving voice interaction, and making distance measurements more reliable.

Built With

Share this project:

Updates

Submission history