Skip to content

Latest commit

 

History

History
26 lines (17 loc) · 1.02 KB

README.md

File metadata and controls

26 lines (17 loc) · 1.02 KB

WhyteSyght

Project Inforamtion

Creating an interface for the blind people which will help them in their day to day life, The system will be wireless and offline based.

Challenges faced

Training action detection algorithm in a such a short time and getting its accuracy so that its usable

Syncing hardware buttons with software so that a smooth process is enabled

Keeping the product as simple as possible while not losing on any basics that is keeping the product simple so that it is easily used by a blind person and keeping features so that it helps him/her in his daily routine.

Execution and Contribution

Naman Bansal @psifrous - Using MQTT protocol to integrate physical buttons & setting up protocol for live stream using Nginx engine

Ambuje Gupta @ambuje- Action Recognition using TensorFlow and OpenCV

Harsh Kataria @rockangator - Scene Detection using TensorFlow and OpenCV

Technologies used:

Raspberry Pi
Node-MCU

Libraries

Tenserflow
Open
CV
OS
Numpy
sys