• DentalX | Deep Learning (Computer Vision)

  • Description: Developed visually intelligent Dental X-Ray diagnosis to detect the dental diseases and provide recommendations to the dentist using deep learning and object detection with Flask APIs
    Libraries: TensorFlow, OpenCV, XML, pandas
    Tools Used: Jupyter Notebook, Spyder, Google Colab
  • Generate Faces | Computer Vision (GAN)

  • Description: Generated faces from a pair of multi-layer neural networks generator and discriminator that compete against each other until one learns to generate realistic images of faces using CelebFaces Attributes (CelebA) dataset
    Concepts: GAN, DCGAN, Pix2Pix and CycleGAN
  • Automatic Image Captioning | Computer Vision & NLP

  • Description: Combined CNN and RNN knowledge to build a deep learning model produces captions for given input image. CNN transforms an input image into a set of features and RNN that turns features into the rich language
    Concepts: Faster R-CNN, YOLO, Single Shot Detection(SSD), LSTM


  • Facial Keypoint Detection | Computer Vision

  • Description: Used image processing & CNN techniques to detect faces in an image and find positions of facial keypoints like eyes, nose and mouth on a face
    Concepts: Convolutional Neural Network, Image Segmentation
  • TV Scripts Generation | NLP

  • Description: Built own Recurrent Networks and Long Short-Term Memory Networks with PyTorch to perform sentiment analysis and generate new text and used RNN to generate new text that resembles a training set of TV scripts
    Concepts: LSTM, Word2Vec and Sentiment Prediction



  • Emergency Fire Exit System in Bus | IoT and Cloud

  • Tools Used: Raspberry Pi 3, Arduino Uno, ThingSpeak, IBM Cloudant, Android Studio
    Sensors: MQ2 Gas Detector, Temperature and Humidity, Servo Motor, GSM900A, GPS Sensor, 16x2 LCD Display
  • KarConnect | IoT

  • Description: IoT based real time vehicle diagnosis and driving behavior monitoring platform
    Tools Used: Raspberry Pi 3, Arduino Uno and Mega, On Board Diagnostics (OBD-II), OBD Simulator, Automosys Dongle Simulator
    Sensors: PIR Motion Sensor, Ultrasonic Sensor, Servo Motor, DHT11, HC05, 16x2 LCD Display


  • Funding Stack | J2EE and IBM DB2

  • Tools Used: Eclipse Neon, BootStrap, IBM Rational Application Developer, IBM DB2
    Modules: Live chat between Innovator and Sponsor, Image Handling, Mail API, Database Administrator, Front-End Development
  • Apps Control Docker | Python (Linux)

  • Tools Used: Python GUI, Tkinter, Ubuntu v16.04, Inkscape
    Modules: Ubuntu Docker, Customization, Access Operating System


Innovation | Research

Movable Seat for Physically Disabled in Cars

This innovation is to have movable seats in four-wheeler so that they can be rotated on their axis and pulled outside. Then the person can sit on it easily, push the seat inside the car and rotate back to the normal position. An objective of this innovation is to provide sliding and rotating seat for a vehicle which aids in ease of getting into and getting out of a vehicle for a needy person; which will facilitate to push-pull, back-forward, up-down and rotational movements of the seat. Such a seat may be useful for a physically disabled person or an obese person.


An Angularly and Laterally Displaceable Seat for Vehicles

Patent Application No.: 3548/MUM/2014
Publication Authority: Intellectual Property India
Research Project: Movable Seat for Physically Disabled in Cars
Checkout code of all Projects at   GitHub