I am a PhD CSE student at the Paul G. Allen School of Computer Science and Engineering at University of Washington, Seattle (UW), working with my advisor Prof. Jon Froehlich and in close collaboration with Prof. Jeffrey Heer. My research interests are at the intersection of HCI, data visualization, urban technology, and civic engagement, and their application in domains such as accessibility and sustainability. In my PhD, my main focus is on urban accessibility. I am working on building tools that enable understanding of, and serve towards advocacy and policy change for urban accessibility. To do this, I am reappriopriating online street-view imagery to assess sidewalk accessibility (Project Sidewalk) and utilizing data visualization techniques to communicate about urban accessibility.
I was in the Computer Science department of University of Maryland, College Park (UMD) for the first two years of my PhD. At UMD, I worked in the Human Computer Interaction Lab (HCIL). Prior to joining PhD, I was a research scholar at IIIT-Delhi. I worked with Prof. Amarjeet Singh in the Mobile and Ubiquitous Computing (MUC) research group and also worked in collaboration with Prof. Yuvraj Agarwal (CMU) and Prof. Anind Dey (UW, then CMU) on my research.
|Jul'19||Some non-academic news: Started a new personal music project "Suroporna" for pursuing my lifelong passion for singing. Check out my Youtube channel and listen to my covers!|
|Jul'19||Won the Amazon Catalyst Award for $10K for funding my thesis work on interactive tools for assessing urban accessibility!|
|Jun'19||Paper on my internship work with Microsoft Research on last-few-meters wayfinding problem accepted at ASSETS 2019! Off to Pittsburgh in October 2019!|
|May'19||Featured on a Seattle local news channel, Kiro 7 News talking about Project Sidewalk!|
|Mar'19||CHI 2019 paper on Project Sidewalk won the Best Paper Award!|
|Dec'18||Paper on Project Sidewalk accepted in CHI 2019. Going to Glasgow, UK to present it! :)|
|Oct'18||Poster presentation and demo of our interactive tool to visualize physical accessibility in ASSETS 2018.|
|Jun'18||Started a research internship in Microsoft Research, Redmond, USA with Enable and Ability Group working with people with visual impairments.|
|Apr'18||Student Volunteering at CHI'18 in Montreal, Canada.|
|Oct'17||Presented and demoed Project Sidewalk in ASSETS 2017 in Baltimore, MD.|
|Sep'17||Moved to University of Washington, Seattle at the Paul G. Allen School of Computer Science and Engineering for pursuing PhD in Computer Science and Engineering.|
|Feb'17||Patent filed on the work done during Summer 2016 internship at Adobe Research, San Jose, CA.|
|Feb'17||Presented a talk on Project Sidewalk at the WalkHackNight II organized by Transportation Techies in DC.|
|Jan'17||Became grad student lead of Project Sidewalk.|
|Dec'16||Paper on Novice Thermography accepted in CHI 2017!|
|Aug'16||Got RAship for Crowd Powered Street-Level Accessibility project with Prof. Jon Froehlich!|
|May'16||Started internship at Adobe Research with the Big Data Experience Lab!|
|Mar'16||Awarded ACM-W Scholarship 2016 for attending CHI 2016.|
|Feb'16||Paper accepted in CHI 2016 Workshop on Future of Human-Building Interaction.|
My research interests are broadly in the areas of Human Computer Interaction (HCI) and Urban Informatics. I enjoy multidisciplinary application-oriented research that involves designing, developing, and evaluating systems in the real world, and that can be applied to solve high value social problems. My current and past work falls within these sub-domains: accessibility, sustainability, data visualization, ubiquitous computing, and smart spaces. I also have an interest to work in ICTD or technology for development.
Ongoing Research Projects
Interactive Visual Exploration of Urban Accessibility at Scale
Advisor: Jon Froehlich (UW) (January 2018 - present)
Collaborator: Jeff Heer (UW)
This work focuses on utilizing accessibility data sources to build novel interactive visualization tools. I am leading the research and development efforts for building an interactive system that provides an at-a-glance visualization of urban accessibility using data collected from Project Sidewalk and other city data sources. As part of this effort, we are also developing a personalizable accessibility model that can assess and score accessibility of regions based on specific user preferences and/or needs of mobility impaired users (e.g., manual wheelchair user vs powered wheelchair user).
Paper(s): ASSETS 2018 Poster
Project Sidewalk: Enabling Crowd-powered Street-level Accessibility Data Collection · ·
Advisor: Jon Froehlich (UW) (August 2016 - present)
Accessible cities are essential for a truly inclusive society. This project looks into scalable methods for collecting and using accessibility data for creating useful Assistive Location-based Technologies (ALTs) to support people with mobility impairments. In particular, we are using crowdsourcing techniques to collect street level accessibility data about cities. We want to build novel applications that will aid people with mobility impairments to plan their travel and navigate efficiently in cities, and also, help other stakeholders such as city governments to determine accessibility of cities with the help of interactive visualization tools.
Paper(s): CHI 2019 · ASSETS 2017 Poster
Past Research Projects
Designing for the Last-Few-Meters Wayfinding Problem for People with Visual Impairments · ·
(June 2018 - July 2019)
Mentors: Alex Fiannaca, Meredith Ringel Morris, Ed Cutrell, Melanie Kneisel
This work was done during a summer at Microsoft Research, Redmond. In this project, we studied the problem of navigating the last few meters of a destination when GPS technologies fail for people with visual impairments. Current GPS systems brings a user to the vicinity of a destination but not to the exact location, causing challenges such as difficulty locating building entrances or a specific storefront from a series of stores. We studied this problem space in two studies: (1) the formative study aimed at understanding challenges, current resolution techniques, and user needs in this navigational context; and (2) in the design probe study, we developed and used a vision-based system called Landmark AI that provided different forms of information; we studied their utility in addressing some of the challenges in the last few meters. Based on these investigations, we articulated a design space for systems addressing this challenge, along with implications for future systems to support precise navigation for people with visual impairments.
Paper(s): ASSETS 2019
Exploring the Role of Thermography in Creating Novel Human-Building Interactions
Advisor: Jon Froehlich (UMD) (September 2015 - January 2017)
Collaborator: Matthew L. Mauriello (PhD Student, UMD)
In this project, we investigated the potential to engage the public in new Human-Building Interactions by expanding their ability to: perform energy audits, survey public infrastructure, and contribute to urban energy analysis.
Paper(s): CHI 2017 · CHI 2016 Workshop
Personal Energy Monitoring in Smart Homes
Advisor: Amarjeet Singh (IIIT-Delhi) (July 2013 - July 2015)
Collaborators: Yuvraj Agarwal (CMU) and Anind Dey (CMU)
This goal of this project was to explore the use of smartphone sensors and electricity meter for inferring daily activities and identifying the individuals who performed them. The results of this study is published in the EnergyLens paper (ACM e-Energy 2014). The system architecture was evaluated for shared living spaces and a room-level energy apportionment system called WattShare (ACM BuildSys 2014) was built. It was part of the larger HumanSense Project.
I worked on building an end-to-end energy apportionment and feedback system for smart homes with an objective of providing occupants with relevant information about their personal energy consumption at the right time. It is based on the original EnergyLens system.
Paper(s): ACM e-Energy 2014 · ACM BuildSys 2014
Personal Activity Monitoring in Shared Spaces
Advisor: Amarjeet Singh (IIIT-Delhi)(March 2014 - October 2014)
Collaborator: Kumar Padmanabh (Bosch Labs, Bangalore)
This project aimed to develop insights into personal activity monitoring in shared spaces, specifically in the home environment, through the use of pervasive computing technologies. In particular, we wanted to leverage the smartphone, a rich source of information, to develop insights for personal activity monitoring and feedback services for the DIY sector. The objective was to do so in the context of using power tools in home environments.
SensorAct: An Occupant-centric Middleware for Buildings
Advisor: Amarjeet Singh (IIIT-Delhi) (November 2012 - May 2013)
Collaborator: Mani B. Srivastava (UCLA)
As part of the Energy Team in IIIT-D, I worked on SensorAct, a middleware for building management. It interfaces with diverse building sub-systems and sensing systems; makes the data available and allows sharing of data and control with users in a secure manner keeping in mind privacy issues.
Paper(s): IEEE UIC 2015 · NSDI 2013
Bandwidth Management Framework for Multicasting over Wireless Mesh Networks
Advisor: P. V. Krishna (VIT University, Vellore)(September 2011 - May 2012)
As a Masters student, I took up an independent research project wherein I developed a bandwidth management framework for performing multicasting over wireless mesh networks. This was later converted into my Masters Project.
Paper(s): IJIEE 2012
Project Sidewalk: A Web-based Crowdsourcing Tool for Collecting Sidewalk Accessibility Data At Scale
Manaswi Saha, Michael Saugstad, Hanuma Maddali, Aileen Zeng, Ryan Holland, Steven Bower, Aditya Dash, Sage Chen, Anthony Li, Kotara Hara, Jon E. Froehlich
CHI 2019: SIGCHI Conference on Human Factors in Computing Systems
Full Paper · Best Paper Award · Acceptance Rate: 23.8%
Manaswi Saha, Kotaro Hara, Soheil Behnezhad, Anthony Li, Michael Saugstad, Hanuma Maddali, Sage Chen, Jon E. Froehlich
ASSETS 2017: International ACM SIGACCESS Conference on Computers and Accessibility
Pandarasamy Arjunan, Manaswi Saha, Haksoo Choi, Manoj Gulati, Amarjeet Singh, Pushpendra Singh, Mani B. Srivastava
IEEE UIC 2015: IEEE International Conference on Ubiquitous Intelligence and Computing
Full Paper · Acceptance Rate: 30.6%