Skip to content

VisioMate

VisioMate, powered by computer vision, works in a similar way to help visually impaired people navigate obstacles. It captures objects in real time and guides users by telling them what is in front of them.

Code Demo lablab.ai

Core trick

Runs real-time object detection on camera frames and announces obstacles to the user.

Tags

accessibilityvision

Description

Are you visually impaired? Do you feel left out? Do you find it difficult to navigate? Do you always have to wait for someone to help you? Do you feel sad when you bump into something and people laugh at you? When you’re crossing the road and see a car coming toward you, you can instantly estimate how far away it is. This quick, almost instinctual judgment comes from your spatial understanding of the surroundings. Based on this sense, you can decide whether to speed up, stop, or keep walking. VisioMate, powered by computer vision, works in a similar way to help visually impaired people navigate obstacles. It captures objects in real time and guides users by telling them what is in front of them.
Kashif581/VisioMate_LabLab.ai_CO-Creating_With_GPT-5_Hackthone★ 00 forksPythonCreated Aug 24, 2025 · last push Aug 26, 2025

README excerpt


* Are you visually impaired?
* Do you feel left out?
* Do you find it difficult to navigate?
* Do you always have to wait for someone to help you?
* Do you feel sad when you bump into something and people laugh at you?

### VisioMate
---
When you’re crossing the road and see a car coming toward you, you can instantly estimate how far away it is. This quick, almost instinctual judgment comes from your spatial understanding of the surroundings. Based on this sense, you can decide whether to speed up, stop, or keep walking.
VisioMate, powered by computer vision, works in a similar way to help visually impaired people navigate obstacles. It captures objects in real time and guides users by telling them what is in front of them.

---

### WorkFlow
---
<img width="996" height="501" alt="newst" src="https://github.com/user-attachments/assets/28a283c2-698b-4a8e-969b-d9586b48dbb5" />

### Tech Stack

* OpenCV (for video processing)
* YOLOv8 (for real-time object detection)
* Gemini (for natural response)
* Pyttsx3 (for speech)
* Streamlit (UI)



https://github.com/user-attachments/assets/d4e671b4-6120-4c18-b1d6-034613e19df8



  [Presentation](https://www.canva.com/design/DAGwgwYtczs/GXBbrDj50jQ1JkwTyVsmwA/view?utm_content=DAGwgwYtczs&utm_campaign=designshare&utm_medium=link2&utm_source=uniquelinks&utlId=ha7abbf2bdf)

Team