Find all the codes to run and build all the features of 6th Sense technologies.
Our Product The Sixth-Sense will serve as an interface cum assistant for the specially-abled individuals primarily suffering from visual impairment, hearing impairment and muteness.6th sense aims to bridge the gap between such individuals and the real world along with its challenges upto a level that hasn’t been achieved yet.
To get a local copy up and running follow these simple steps.
Run the below code in the terminal of your environment.
pip install -r requirements.txt Follow the Below Instructiins to use the codes as features of this repo.
The folder Hand_symbol_Detection contains the code for the same.
Epoch.pt contains the trained model weights to run the model.
Run Main.py to get the hand symbol detected output from a given hand image.
The folder Pose_Estimation_Mobilenet contains the code for the same.
utils folder contains the training utils to run and detect body pose.
weights folder ontains the trained model weights to run the model.
Model.py contains the code for model architrecture
Run Video.py to get the pose estimation and a roi around the hand which coould be used for hand symbol recognition.
The GUI has 4 parts:
- Hand Symbol Recognition
- OCR
- Image Captioning
- Text to Speech
To run the gui prepared in PyQt5
cd DashBoard python main.py6thSense.mp4
text2speech_and_speech2text.ipnyb contains the code for speech and text conversion. This is to help the dumb people to convert their words into speech. They need to write what they want to say in the text field of the software and click on play generated speech button. Then their words will be converted into speech that will help them to communicate.
Model.py contains the code for the ocr to get characters from an image Optical character recognition convert the text present in image to machine encoded text. This is to help the blind people to read. To help the blind people we have converted the generated text to speech. This also helps the dumb people. They can write whatever they wanted to speak in a paper and upload the image. Then our software will convert that to speech. To use the ocr you need to click the upload image button ,then choose an image location and then click open. Then our software will convert the text present in image to speech.
This software captions the image and helps the blind person to see the world. The captioned text is converted to speech to help the blind person. We can upload image in two ways.One by selecting the image location and the other one by taking picture directly from the camera. To upload a image select upload an image button and select an image location and click ok. After that our software will caption the image and convert it to text. To click an image from camera, click on the click a picture button and then click on click photo andd close the camera tab. Then our software will caption the clicked image and convert it to speech.
Distributed under the MIT License. See LICENSE for more information.
Kamaljeet Sahoo- KamaljeetSahoo
Smruti Sikha Panigrahi - SmrutiSikha
Komal Trishna Behera - komalbehera
Raj Aaryaman Patra - RajArPatra