Big Data-Driven Pedestrian Analytics: Unsupervised Clustering and Relational Query Based on Tencent Street View Photographs

Abstract

Recent technological advancements in geomatics and mobile sensing have led to various urban big data, such as Tencent street view (TSV) photographs; yet, the urban objects in the big dataset have hitherto been inadequately exploited. This paper aims to propose a pedestrian analytics approach named vectors of uncountable and countable objects for clustering and analysis (VUCCA) for processing 530,000 TSV photographs of Hong Kong Island. First, VUCCA transductively adopts two pre-trained deep models to TSV photographs for extracting pedestrians and surrounding pixels into generalizable semantic vectors of features, including uncountable objects such as vegetation, sky, paved pedestrian path, and guardrail and countable objects such as cars, trucks, pedestrians, city animals, and traffic lights. Then, the extracted pedestrians are semantically clustered using the vectors, e.g., for understanding where they usually stand. Third, pedestrians are semantically indexed using relations and activities (e.g., walking behind a guardrail, road-crossing, carrying a backpack, or walking a pet) for queries of unstructured photographic instances or natural language clauses. The experiment results showed that the pedestrians detected in the TSV photographs were successfully clustered into meaningful groups and indexed by the semantic vectors. The presented VUCCA can enrich eye-level urban features into computational semantic vectors for pedestrians to enable smart city research in urban geography, urban planning, real estate, transportation, conservation, and other disciplines.

Spatially Varying Effects of Street Greenery on Walking Time of Older Adults

“…The built environment is a known and extensively studied concept in fields such as geography, environmental science, transportation, public health, and urban planning [10]. Recently, with the advancement of science and technology (e.g., compute vision as well as deep learning algorithms) [11][12][13], street greenery, a built environment factor that cannot be easily assessed by traditional approaches, can now be accurately calculated based on street view imagery (e.g., Google, Baidu, and Tencent) data, which effectively describe real-world scenery [14,15]. Therefore, it has recently attracted scholarly attention.…”

Role of the Built Environment in the Recovery From COVID-19: Evidence From a GIS-Based Natural Experiment on the City Blocks in Wuhan, China

“…In the future, we look forward to seeing studies on BECs associations and impacts on different cities and countries with different intervention conditions so that the findings can be further verified thoroughly. More quantified urban data sources, such as city information modeling (Xue, Wu et al, 2021) and computable street view features (Xue, Li et al, 2021), might reveal new evidence as well. Encouragements and subsidies based on solid evidence, for example, to increase the quantities of nearby hospitals, promote residential landscape sanitary management levels, and take prioritized measures for targeted economic city blocks should be further recommended for sustainable and smart urban development.…”