Case Studies
By Sector
Advanced Search
By Sector
Advanced Search
LetsTalkSign - Communication made easier for hearing imapaired
Problem: Hearing impaired people face numerous challenges in their daily lives including access to education. There is a need for a solution that can aid people who face a difficulty/disability in hearing and communicate via sign language.
Solution: The solution leverages artificial Intelligence-powered to enable easy two-way communication between those hearing-impaired users who use sign language for communication and others who do not know sign language.
Lipread - An AI powered solution for people with speaking diabilities
Problem: There is a need for a solution that can add voice to the videos wherever needed and help the patients suffering from speaking disabilities to be able to express themselves through their lip movements
Solution: Visual Speech Recognition technology is implemented by tracking and extracting the movement of the subject’s lips, using Deep Neural Network (DNN) techniques to analyse the lip movement. The results are compared with a universal model to determine what has been spoken.
Skywise Digital Solution is using historical data to deal with Aircraft maintenance related problems
Problem: Some of the worst aircraft disasters have been attributed to faulty or overlooked maintenance. Maintenance is one of the top five causes of domestic aircraft delays. The objective of this application is to enhance the reliability aircraft maintenance through maintenance prediction.
Solution: The application uses historical maintenance information from aircraft and fleets along with thousands of data parameters collected in real time from each aircraft in an airline’s fleet.
A simple 'Plug and Play' AI solution is speeding up the baggage security
Problem: Baggage handling and its security is a major operational challenge at airports. Manual detection of prohibited items in X-Ray images of scanned baggage is a cumbersome task that is highly prone to human error and often leads to a security breach.
Solution: An easy-to-use “Plug and Play” module which can be connected to the baggage scanner to detect the harmful/banned objects with the help of AI and show them prominently with the help of shades and colour.
Cameras to capture Shelf condition in Real-time to maintain Stock
Problem: There is a need for a technological platform that can help organizations and retail businesses to organize their shelf inventory in an efficient manner to avoid errors and revenue loss because inaccurate inventory.
Solution: Real-time shelf condition is captured using in-store cameras and analyzed to identify status at Stock Keeping Unit (SKU) level. Images are sent to cloud servers, where deep-learning algorithms that identify each Stock Keeping Unit. Each store’s camera network is tailormade for its size and layout. Shelf-edge cameras are great for main aisles, dome cameras for fresh categories, and autonomous robots for new products, that provide the data to be analyzed by the ML and DL algorithms to provide stock level information.
AI will Assist Uber users in payment
Problem: The major inconvenience for employees who are using Uber for business trips is the constant mix up between personal profiles and business profiles. As a result, employees get accidentally charged for their business rides to their personal account and vice-versa. This happens because of the rider forgets to change their payment method
Solution: Uber uses AI and machine learning to identify the ride type according to the ride details entered by the user and suggests the ride account to the employees while making a payment. With deep networks and decision trees in machine learning, the company is now able to predict the payment method of a customer even if they forget to change it by displaying a pop-up during ride booking.
AI will decide what you wear
Problem: Predicting fashion trends is a key challenge faced by the ecommerce sector. Despite having a large amount of data, companies find it difficult to understand customer preferences.
Solution: The platform scans fashion portals, social media, and Myntra customer data and uses computer vision and machine learning to understand customers preferences. Data on whether customers are buying products or not are fed into a computer, which continues to learn and throw up what works best for customers.
AI is helping Swiggy know what you want to eat
Problem: Swiggy seeks to make its app more efficient by providing a personalized experience for each user. Swiggy aims to provide a prediction model that helps restaurants plan for high demand periods.
Solution: Machine Learning is used to generate a personalized list of restaurants for consumers based on their past orders, searches and interactions. The algorithm enables faster discovery of restaurants by analyzing thousands of data points such as food preferences, a live snapshot of delivery partners in consumer’s area, etc.
Smart Vending Machines
Problem: Customers seek convenience coupled with healthy food options. This has increased the demand for smart vending machines that display calorie counts, provide hot or cold delivery and offer several payment options.
The objective is to design a smart vending machine leveraging the use of AI for a better customer experience
Solution: The Cloud Vending machine uses computer vision and facial recognition technology to identify customers and merchandise. This smart vending machine uses AI, IoT sensors and deep learning capabilities to provide product recommendations, targeted advertising, smart inventory management and easy integration with other business systems.