Marketing ads often misrepresent themselves and cater to the audience in an unfair manner. We make sure our model exploring the consumer's personality in recommending ads addresses this concern.
Apply adversarial noise to profile photos to combat online surveillance and facial recognition tools.
RAITE is a testing environment for ML apps to obtain scientifically unbiased results through use of special Tools and Workflows for Responsible AI assessments.
Create animations using an autoencoder trained on motion capture data
AI powered personalized fitness expert for working-out from Home
A unique ROS package for accelerating ML model development using ROS simulated data
It analyzes what is said on a Youtube Video using the Toxicity classifier build on tfjs
Want to create a custom Google Doodle? Just enter your subject and event, and this web app will use dynamic AI techniques to create a unique Google Doodle for you.
Combating global pandemics with data
We built a digital stethoscope for under $1, using it with AI to diagnosing respiratory symptoms. Our goals is to empower doctors to treat more patients via Telemedicine for COVID19.
Put the brakes on disease outbreaks with artificial intelligence.
A new NLP based search engine for COVID19, biology and medicine related scientific papers
Predicting the price of new supermarket products
This project is a new type of research and an unprecedented global project that connects scientists, researchers and inventors in a way that has never existed before.
An autonomous drone delivery system to provide efficient and virus-free deliveries.
Reopening Economy In the Times of Corona
One-stop platform for the citizens to access covid-related services and help from official government sources.
Using Electrical Impedance Tomography to create a portable, (much) cheaper and instantaneous lung scanning band to help doctors monitor patients without human interaction
Stay home. Stay informed.
A machine learning API that makes building and deploying models as simple as a one-line function call.
Facial recognition algorithms are often biased because the training data doesn't contain enough diverse faces. The solution uses StyleGAN2 to create additional samples to train models on
A Web Application Prototype built for Doctors to find out whom to test for the infection first under a limited testing capacity by finding out the probability of a person having the infection.
Perform feature extraction by constructing graph structure representing the molecules and calculating 1D 2D and 3D descriptors for the molecules.Our aim is to develop a model to detect the HIV drugs.
1 – 24 of 24