Project Sidewalk
Project Description
This project describes a two-pronged vision: first, to develop scalable data collection methods for acquiring sidewalk accessibility information using a combination of crowdsourcing, computer vision, and online map imagery, and second, to use this new data to design, develop, and evaluate a novel set of navigation and map tools for accessibility. Our overarching goal is to transform the ways in which accessibility information is collected and visualized for every sidewalk, street, and building façade in America.
Publications
Accessibility for Whom? Perceptions of Mobility Barriers Across Disability Groups and Implications for Designing Personalized Maps
Proceedings of CHI 2025 | Acceptance Rate: 24.9% (1249 / 5020)
PDF arXiv doi Cite Code Project Sidewalk Accessibility for Whom?
Towards Fine-Grained Sidewalk Accessibility Assessment with Deep Learning: Initial Benchmarks and an Open Dataset
Extended Abstract Proceedings of ASSETS 2024 | Acceptance Rate: 66.7% (58 / 87)
PDF doi Cite Deep Learning for Sidewalk Assessment Project Sidewalk
“I never realized sidewalks were a big deal”: A Case Study of a Community-Driven Sidewalk Audit Using Project Sidewalk
Proceedings of CHI 2024 | Acceptance Rate: 26.3% (1060 / 4028)
PDF doi Cite Project Sidewalk + Community Science Project Sidewalk
LabelAId: Just-in-time AI Interventions for Improving Human Labeling Quality and Domain Knowledge in Crowdsourcing Systems
Proceedings of CHI 2024 | Acceptance Rate: 26.3% (1060 / 4028)
Implementing a Community-Based Virtual Tool To Characterize Sidewalk Accessibility in a Northern New Jersey (NJ) Town
Extended Abstract Proceedings of ASCIP 2023 Annual Meeting 2023
Implementing a Community-Based Virtual Tool To Characterize Sidewalk Accessibility in a Northern New Jersey (NJ) Town
Extended Abstract Proceedings of CMSC 2023 Annual Meeting 2023
Scaling Crowd+AI Sidewalk Accessibility Assessments: Initial Experiments Examining Label Quality and Cross-city Training on Performance
Proceedings of ASSETS'22 2022 | Acceptance Rate: 58.9% (43 / 73)
PDF doi Cite Project Sidewalk Deep Learning for Sidewalk Assessment
Participatory Design of Crowd+AI Tools to Map, Analyze, and Visualize Sidewalk Accessibility
Extended Abstract Proceedings of TRANSED 2022
Towards Global-Scale Crowd+AI Techniques to Map and Assess Sidewalks for People with Disabilities
CVPR2022 Workshop: Accessibility, Vision, and Autonomy (AVA) 2022
PDF arXiv doi Cite Project Sidewalk Deep Learning for Sidewalk Assessment
Visualizing Urban Accessibility: Investigating Multi-Stakeholder Perspectives through a Map-based Design Probe Study
Proceedings of CHI 2022 | Acceptance Rate: 24.7% (637 / 2579)
PDF doi Cite Talk Project Sidewalk AccessVis Accessibility-Infused Maps Multi-Stakeholder Interviews to Examine Urban Accessibility
Experimental Crowd+AI Approaches to Track Accessibility Features in Sidewalk Intersections Over Time
Proceedings of ASSETS 2021 | Acceptance Rate: 61.8% (55 / 89)
Sidewalk Gallery: An Interactive, Filterable Image Gallery of Over 500,000 Sidewalk Accessibility Problems
Proceedings of ASSETS 2021 | Acceptance Rate: 61.8% (55 / 89) | Best Artifact Runner-up Award
Sidewalk Accessibility in the US and Mexico: Policies, Tools, and A Preliminary Case Study
Extended Abstract Proceedings of Civic Tech 2020
Towards Mapping and Assessing Sidewalk Accessibility Across Socio-cultural and Geographic Contexts
Extended Abstract Proceedings of Data4Good 2020
Deep Learning for Automatically Detecting Sidewalk Accessibility Problems Using Streetscape Imagery
Proceedings of ASSETS 2019 | Acceptance Rate: 25.9% (41 / 158) | Best Paper Award
PDF doi Cite Code Project Sidewalk Deep Learning for Sidewalk Assessment
Project Sidewalk: A Web-based Crowdsourcing Tool for Collecting Sidewalk Accessibility Data at Scale
Proceedings of CHI 2019 | Acceptance Rate: 23.8% (705 / 2960) | Best Paper Award
Grand challenges in accessible maps
Interactions 2019
A Feasibility Study of Using Google Street View and Computer Vision to Track the Evolution of Urban Accessibility
Extended Abstract Proceedings of ASSETS 2018
Interactively Modeling and Visualizing Neighborhood Accessibility at Scale: An Initial Study of Washington DC
Extended Abstract Proceedings of ASSETS 2018
PDF doi Cite Project Sidewalk Accessibility-Infused Maps AccessVis
SIG: Making Maps Accessible and Putting Accessibility in Maps
Extended Abstract Proceedings of CHI 2018
A Pilot Deployment of an Online Tool for Large-Scale Virtual Auditing of Urban Accessibility
Extended Abstract Proceedings of ASSETS 2017
Scalable Methods to Collect and Visualize Sidewalk Accessibility Data for People with Mobility Impairments
UMD CS PhD Dissertation 2016
Temporal Tracking Urban Areas using Google Street View
UMD CS MS Thesis 2016
The Design of Assistive Location-based Technologies for People with Ambulatory Disabilities: A Formative Study
Proceedings of CHI 2016 | Acceptance Rate: 25.0% (600 / 2400)
PDF doi Cite Accessibility-Infused Maps Project Sidewalk Multi-Stakeholder Interviews to Examine Urban Accessibility
Scalable methods to collect and visualize sidewalk accessibility data for people with mobility impairments
Extended Abstract Proceedings of UIST 2014
An Initial Study of Automatic Curb Ramp Detection with Crowdsourced Verification using Google Street View Images
Extended Abstract Proceedings of HCOMP 2013
A feasibility study of crowdsourcing and google street view to determine sidewalk accessibility
Extended Abstract Proceedings of ASSETS 2012
Videos
Talks
Project Sidewalk: Crowd+AI Techniques to Map & Assess the World's Sidewalks
Feb 29, 2024 | NSF Smart and Connected Communities Panel on “Pathways to Transitioning Project Outcomes"
Nashville, Tennessee
Scalable Techniques to Study the Equitable Distribution and Condition of US Sidewalks
Dec 14, 2022 | ITS (Intelligent Transportation Systems)
Tacoma, Washington
Crowd + AI Tools for Scalable Sidewalk Assessment
Dec 13, 2021 | Spatial Data Science Symposium 2021
Virtual
Designing Interactive Computational Tools for Understanding Urban Accessibility at Scale
May 28, 2021 | PhD Generals Exam
Virtual
Project Sidewalk: Mapping the Accessibility of the Physical World at Scale using Interactive Computational Tools
Jan 22, 2020 | Google Tech Talk
Seattle, WA
Project Sidewalk: Mapping the Accessibility of the Physical World at Scale using Interactive Computational Tools
Sep 30, 2019 | WSRO 2019 Keynote
Spokane, WA
Project Sidewalk: A Web-based Crowdsourcing Tool for Collecting Sidewalk Accessibility Data at Scale
May 07, 2019 | CHI'19
Glasgow, UK
Project Sidewalk: Mapping the Accessibility of the World through Google Street View
Oct 09, 2018 | Change Seminar
University of Washington, Seattle
Bridging the Pedestrian Accessibility Informational Gap: User-Facing Applications and Large-Scale Virtual Auditing
Nov 15, 2017 | UW CSE Affiliates 2017
Seattle, WA
Project Sidewalk: Assessing Urban Accessibility using Crowdsourcing and Google Street View
May 25, 2017 | HCIL Symposium 2017
College Park, Maryland
Making with a Social Purpose
Apr 06, 2017 | Lecture Series at the Laboratory for Telecommunication Sciences
LTS Auditorium, College Park, MD
Project Sidewalk: Characterizing Physical World Accessibility At Scale
Feb 08, 2017 | WalkHackNight II
Arlington, Virginia
Interactive Computational Tools for Accessibility
Nov 07, 2016 | Diversity in Computing Summit 2016
College Park, Maryland
Uran Oh, Ladan Najafizadeh, Meethu Malu, Manaswi Saha, Lee Stearns
Tech+Design: Interaction Design for a Purpose
Nov 03, 2016 | Technica: Tech+X Talk Series
University of Maryland, College Park
Soheil Behnezhad, Majeed Kazemitabaar, Manaswi Saha, Liang He
Characterizing Physical World Accessibility at Scale
Nov 03, 2016 | GroupSight @ HCOMP2016
Austin, Texas