With the aid of the numerous effective vaccines, the world is now sprinting fast towards the end of the pandemic. The year 2020 unravelled several fractures in the way we function. It was because we could interact and communicate effectively that the whole of this alienated year elapsed.
Researchers, all around the world, are coming up with their most revolutionary works to alleviate the global status of well-being. And the pandemic has only accelerated this progress further. Among these pioneers is Isha Chaturvedi, a data scientist consistently leading the breakthroughs by creating effective solutions to extremely diverse, complex problems.
Bagging her Environmental Technology and Computer Science undergraduate degree from Hong Kong University of Science and Technology (HKUST) and her Data Science master's degree from New York University (NYU), Isha has always immersed herself in activities that extended beyond the classrooms. She excels as a data scientist, and her work at several organisations and in projects demonstrates an unmatched prowess.
Her astounding work as the first author of the research, carried out in collaboration with two other postdoctoral researchers, dedicated to the building of a direction-based navigation app for smart glass wearers for local settings was published in ACM Intelligent User Interface (IUI)
ACM IUI 2019 is the 24th annual international conference of the intelligent interfaces community, and serves as a premier global forum for reporting outstanding research and development on intelligent user interfaces. It is where the Human-Computer Interaction (HCI) community meets the Artificial Intelligence (AI) community.
Beginning with the Research
Upon the completion of her undergraduate study, she began serving as a research assistant under her university Professor Pan Hui at Deutsche Telekon (DT) System and Media Laboratory (SyMLab) - in close cooperation with Telekon Innovation Laboratories (T-Labs). He is a renowned computer scientist and professor at the University of Helsinki and the Hong Kong University of Science and Technology (HKUST). He has won numerous global-level, prestigious honours and accolades, including getting elected as an International Fellow of the Royal Academy of Engineering in the year of 2020, a Fellow of the Institute of Electrical and Electronics Engineers, a Member of the Academia Europaea, and a Distinguished Scientist of the Association for Computing Machinery (ACM). He has also been elected to the endowed professorship Nokia Chair in Data Science.
Aligning the interests of the lab with hers, Chaturvedi devoted her time working on machine learning at SyMLab, developing a direction-based navigation app for localised environments, particularly for the indoor environment for smart glasses like Google Glass, using Wi-Fi localisation techniques. The objective was to design indoor navigation using such smart glasses and improving user experience and interaction with the smart glass. An extensive study on human peripheral vision was conducted.
The Challenge Encountered
The most profound challenge encountered during this research was how smart glasses could lead to the obstructed human central vision - making it tougher to multi-task, thereby, increasing the chances of accidents while wearing them. Chaturvedi unravelled the solution in the peripheral vision of the human eye which could be used for navigation.
Remarkability of the App
Most smart glasses possess a small and limited field of view, restricting their potential. It compels users to direct their central eye gaze towards the small screen of the glass to extract meaningful information, causing visual fatigue and reducing their ability to multi-task. Chaturvedi, guided by her two post-doctoral researchers, intricately explored these shortcomings and the ways to improvise user experience. She modelled her work by introducing mobile peripheral vision (MPV), which can be employed on any smart glass with a head-mounted display without any additional hardware necessities, and ran innumerable usability tests. She, then, created an application for indoor and outdoor navigation - demonstrating its holistic application. Furthermore, Chaturvedi had to conduct her tests in Hong Kong during nighttime to ensure minimum disturbance from cell networks.
Chaturvedi's Previous Works
Chaturvedi employs her exceptional data science skills to curate smart solutions for complex problems. She places empowerment and impact at the centre of all her endeavours.
Her project, Automated Detection of Street-Level Tobacco Advertising Displays, resided in the public health landscape. It targeted the tracking down of vendors that have tobacco advertising displays in an automated way (using data science techniques) to observe how closely located they are from schools, areas of communities of colour, and low-income communities. It was a collaboration project of New York University (NYU) Health and NYU Center of Urban Science and Progress.
She even created a chatbox for Governor's island during her Havas Fellowship. The objective of this project was to come up with a chatbox which aimed at improving tourism at Governor's Island. The chatbox bestowed tourists with great virtual tours and the cinematic displays of the island's history. This designed chatbox was presented at the Library of Congress by Chaturvedi's colleague. The project resembled the idea that the Anne Frank House had employed for their museum.
Furthermore, while working with Ericsson, she and her team developed a proprietary algorithm for geo-localisation and pole health and morphology estimation of Utility Poles and Street Lights. This work of theirs’ targeted at supporting Ericsson 5G network capabilities within the US. The team lead later presented this model at Mobile World Congress, LA.
Chaturvedi's works straddle building intelligent solutions and optimising resources efficiently. Her other participation and engagements boast her stunning experience in drafting and designing meaningful experiences for different communities. She will be continuing to engage in delivering impactful results in the future as well.