Nishal Stanislaus Silva


I have been working as a researcher, both in academia and industry with over a decade of combined experience. Here's a curated list of some of the coolest projects I have done:

Gesture Detection for Electrig Guitar
I conducted initial in

Musical Pattern Detection

Multisensory Concerts

Datasets for Musical Pattern Detection

Technology to Enable Seamless and Continuous Digital Prints of Finished Garments

Continuous Blood Pressure Monitoring for COVID 19 Patients
There once was a time when we were all quarantined at our homes, and the world was in a state of uncertainty. But not everyone was as lucky as me. Many people diagnosed with COVID 19 were in hospitals, and they were in need of constant monitoring. Doctors and nurses, who were already overworked and stressed, while performing their duties commendably, were at the risk of contamination due to prolonged exposure to patients. During this time, I was fortunate enough to be part of a team that worked on a project to develop an IoT system to continuously monitor the blood pressure of COVID 19 patients remotely.

A few of my colleagues and I worked on this project on out own time. Together, but remotely, we designed a system that could continuously monitor the blood pressure and blood oxygen levels of COVID 19 patients.

The entire world was in a state of panic. Procuring specialized equipment was impossible. Our entire solution revolved around retrofitting components to existing medical equipment. We used the internal connections of blood oxygen meters; and webcams with simple image processing algorithms to obtain the measurements. The non invasive equipment could be attached to patients at all times, and the readings were continuously sent to a server which maintained a dashboard for doctors and nurses to monitor the patients.

Automated Quality Checking of Care Labels
Care Labels in clothing provide instructions on how to take care of them - just like the name implies. Most clothes have a care label that is several layers thick, and which has instructions in many languages. The accuracy of the content of such labels is an important factor in determining the quality. Big brands that have high markups always check the integrity of all parts of the garment, including the care label.

As care labels are outsourced, an operator often checks random samples for accuracy. This process can be time-consuming and prone to human error, especially when the user is not familiar with the language or the specific care instructions. To overcome this, I designed a computer vision based system to automatically check the content of each care label against a reference.

The system reduced the time taken for random quality checks by approximately 99.5%, and also reduced the human error. I designed the system to be easily deployable in any factory, and it is easy to learn and use. Figures below show the reference care label, a sample care label, and the results of the system.

Automated Fabric Quality Checking
In an apparel production setting, the raw fabrics must be defect free. Often, entire fabric rolls are checked for defects manually. This step is time consuming, but also limited to only defects that are visible to the human eye. I designed a computer vision based system to automatically check the fabric for defects. The system was designed to be easily deployable in a factory, and it is easy to learn and use. Figures below show several examples of the system. The figures are taken from macro cameras, and feature artifacts and defects that a human operator might potentially miss.

Smart Retail Racks

Smart Prototyping for Garment Prints
One of the coolest projects I've done was an augmented reality table to visualize prints on finished garments. I designed a table with a projector and a camera mounted above it. The camera was used to capture the silouhette of the garments, and the selected design was projected onto the garment. The table was designed to be used in a retail environment, where customers can select a garment and see how it would look with different prints. This project was a great example of how technology can be used to enhance the customer experience in retail.

Below are some photos and a demo video of the project. The project was during my former employment, and I was the lead designer and developer of the system. However, I cannot show the final outcome in a public forum due to Non Disclosure Agreements. The video shows an intial prototype of the system.

Show portfolio