Not everyone understands sign language,this project aims to solve this by removing the need of having a person who understands sign language interpret every single sign.

The project employs the follwing key areas:

1. Computer vision.
2. Deep learning. [TensorFlow library]

1. Recognize simple hand poses by tracing the hand’s joints.
2. Implement TensorFlow object detection.
3. Implement Deep learning & train model.

1. Recognize the flow of movement of both hands finding key areas on a hand. Key-points correspond to the joints.

Dependencies & Tools used

** MediaPipe - Media Pipe is an open source library that enables object tracking by offering customizable M.L solutions for live and streaming media.

** OpenCV Python bindings - OpenCV is an open source framework used for Image and video manipulations i.e Computer vision.

Dependencies - Pre-trained models(cause i am on a laptop)

-- Labelling images for Object Detection - using Label Image package to label images for object detection.[LabelImg] -- Training Tensorflow for Sign Language. -- Detecting Sign language inreal Time.

1. Prepare image labels using labelling & draw boxes over multiple sing language collected with my webcam.
2. Transfer learning using SSD MobileNet.