Projects
INDIAN SIGN LANGUAGE INTERPRETER
Data in sign languages in general is scarce, and in ISL in particular. For the estimated 10 to 20 lakh hearing impaired people in our country, the number of ISL interpreters are in hundreds. This project aims at creating intreprator for ISL.
DATA ANONYMIZATION
Escalation of technology led to an increase in the capability to record and store personal data about consumers and individuals. Data anonymization will allow for the re-use of personal data in privacy- preserving ways and If stolen, it is very hard to re- identify it.
NAVIGATED LEARNING
This project aims at developing a precision learning model. Navigated learning implements precision learning, by computing a variety of semantic embeddings. This project explores Navigated Learning models for enterprise settings.
Clinical Decision Support System
This solution is designed to enable easy integration with external applications. If you are a healthcare application developer and would like to integrate this solution to your application, we expose REST APIs that you can use.
BelYo
A Bloackchain based platform which allows to create digital records like certificates, which will be retrieved by contact tracing application through API’s. Moreover such records retrieved are with user consent and without doubting the authenticity.
GENERALIZED WASSERSTEIN DEEP FOREST
Project aims to design an interpretable, scalable deep model that can handle high-dimensional imbalanced datasets arising in various applied AI domains.
Remotely Accessible lab
Project aims to make a remotely accessible lab that uses hardware on which experiments can be performed. Students can perform experiments through a website where experimental parameters can be configured.
Deep Generative Models
This project aims to developing novel methods which utilize trained deep generative models for improving the deep learning models on diverse geospatial dataset for Earth science applications.
COMMUNITY DISCOURSE MODELING
Project aims at improving Speech to text conversion for Kannada, and develop semantic processing models like topic modeling, argument mining, and sentiment analysis.
CT scan analysis for COVID-19 screening
The goal of the project is to build a system that takes as input CT scans, and answers the following: Covid-infected: Yes or No, along with a confidence score and Contour detection of the lesions.
Automatic Document Processing
Project goal is to generates a structured output, say as an xml format, that gives various information in the document. This can reduce the huge manual effort of entering the data into the system
Privacy preserving ML and neuroscience
A main contribution of this work will be creation and deployment of a privacy preserving machine learning infrastructure for health care, applicable in general.
Medical images segmentation
Accurate segmentation of internal body organs, tissues, lesions etc. from medical images (CT, MRI etc.) are of great value in diagnosis and treatment of various diseases. Given the fact that the world is increasingly seeing newer kinds of pathogens, the need to detect novel types of lesions, tumors etc.
Quantum computing with Nonlinear oscillator networks
As opposed to classical computing, quantum computing (QC) methods employ quantum mechanical phenomena such as interference, superposition, entanglement, etc.
AI Intervention based Scientific Study of Yoga
Due to fast changing lifestyle and imbalanced workload, a considerable percentage of the population suffers with mental problems such as stress, brain related disability, improper sleep etc.
Segmenting Regions and Objects in Images under Minimal Supervision
Automatic segmentation and classification of image regions is relevant in various areas of cognitive computing which involves both natural and medical images.
Connection Setup Mechanisms for Large Quantum Networks
Quantum technologies are going to play an important role in future in multiple dimensions. Quantum computers have started proliferating to solve some of the hardest problems which are difficult for even classical supercomputers.
Modern deep learning to tackle climate data challenges
The classical approach to modeling the dynamics of physical systems relies on intricate physical models that capitalize on robust prior knowledge of underlying phenomena and numerical simulations demanding extensive computations.
Bharatiya Bhasha NLP Research Group
Observing and modeling the dynamics of climate components, such as the atmosphere and ocean, presents significant challenges for scientific analysis and understanding of their evolution. To reduce uncertainties in predicting extreme events and improve the quantification of climate change.
Models for evaluation of artistic outputs in constructivist learning
Learning is a deeply individualized activity, and a given process of learning results in different kinds of assimilation in different learners. One of the ways to understand different patterns of assimilation of a learning material is to use constructivist learning.
3D Mapping of Unstructured Outdoor Environments
This project entails inferences from the data collected using a monocular sensor mounted on an
unmanned robot. This work focuses on outdoor environments, i.e., mixed-built environments.
Community Discourse Modeling using Indic NLP
Janastu is an organization focusing on empowering communities through development of Free and Open Source Software (FOSS) and ICT led interventions.