Designing a User-centric Framework for Information Quality Ranking of Large-scale Street View Images

TL;DR

The paper tackles the challenge of heterogeneous information quality in large-scale street view imagery (SVI) and proposes a user-centric Quality of Information (QoI) framework that decomposes QoI into three attributes: spatial, temporal, and content. It defines metrics for each attribute, integrates them into a unified QoI score, and demonstrates data-driven ranking and querying on a New York City dashcam dataset from Nexar. The contributions include a stakeholder-informed QoI taxonomy, a data-segmentation ranking mechanism to guide data procurement and curation, and practical guidance for building future user-centered, sustainable SVI tools for urban sensing. This framework enables more efficient, equitable, and cost-aware use of SVI in urban planning and research.

Abstract

Street view imagery (SVI), largely captured via outfitted fleets or mounted dashcams in consumer vehicles is a rapidly growing source of geospatial data used in urban sensing and development. These datasets are often collected opportunistically, are massive in size, and vary in quality which limits the scope and extent of their use in urban planning. Thus far there has not been much work to identify the obstacles experienced and tools needed by the users of such datasets. This severely limits the opportunities of using emerging street view images in supporting novel research questions that can improve the quality of urban life. This work includes a formative interview study with 5 expert users of large-scale street view datasets from academia, urban planning, and related professions which identifies novel use cases, challenges, and opportunities to increase the utility of these datasets. Based on the user findings, we present a framework to evaluate the quality of information for street images across three attributes (spatial, temporal, and content) that stakeholders can utilize for estimating the value of a dataset, and to improve it over time for their respective use case. We then present a case study using novel street view images where we evaluate our framework and present practical use cases for users. We discuss the implications for designing future systems to support the collection and use of street view data to assist in sensing and planning the urban environment.

Figures (10)

• Figure 1: A blurry image example from Nexar dashboard camera image dataset (left). The blurriness would lead to low quality of information for many applications that seeks to extract insights from the content of the image (i.e. number of cars or sidewalk width), but can be of sufficient quality for sky view estimation applications.
• Figure 2: Example images from various street view datasets other that Google Street View
• Figure 3: Data distribution of an image dataset collected by vehicles instrumented with dashboard cameras across 196 Zip code areas in New York City (left). Population of each Zip area is shown on right for reference. Note that representation in the dataset does not commensurate the population in most cases.
• Figure 4: Overview of the proposed Quality of Information framework based on spatial, temporal, and content attribute dimension. Such framework can be used to rank, query, and visualize information relevant to specific use cases.
• Figure 5: Spatial data distribution over different time period, a) street network of 14 zip code areas, b) street view data distribution for a single hour, c) a single day (24 hrs), and d) entire data collection period (46 days).