arXiv subjects

Andres Sevtsuk

Publications and source records attributed to Andres Sevtsuk.

4 recordsLinked to original sources

arXiv2026

A Vision-Language Framework for Measuring Social Life on Sidewalks

While a number of methods exist for counting pedestrians in street-view imagery, these mostly ignore the social dimensions of pedestrian activity. A street traversed by a high volume of pedestrians has the same headcount as a street where people linger, sit, and socialize. This paper presents a vision-language framework for extracting social indicators from street-level imagery. Panoramic street-level imagery is reprojected to sidewalk-facing sideviews with preserved timestamps. A vision-language model (VLM)-based activity detection system codes each person across ten independent observable dimensions, resolving a systematic failure mode in which models prompted with high-level social categories conflate observable states with contextual inferences. The resulting social indicator system produces a Social Dwelling Index (SDI) that jointly considers pedestrian grouping and dwelling, provides activity labels documenting behavioral diversity, and issues binary flags for the presence of accessibility-sensitive populations. We apply the framework to 102,514 sideviews in New York City, revealing that pedestrian volume and SDI are only weakly associated (r = 0.168): streets with the highest foot traffic are not where social activity is most intense. The framework provides a scalable method for measuring not only how many people are on city sidewalks, but also their grouping, posture, and activity type, summarizing the non-transient activities that occur on city sidewalks.

cs.CY↗

arXiv2026

MINGLE: VLMs for Semantically Complex Region Detection in Urban Scenes

Understanding group-level social interactions in public spaces is crucial for urban planning, informing the design of socially vibrant and inclusive environments. Detecting such interactions from images involves interpreting subtle visual cues such as relations, proximity, and co-movement - semantically complex signals that go beyond traditional object detection. To address this challenge, we introduce a social group region detection task, which requires inferring and spatially grounding visual regions defined by abstract interpersonal relations. We propose MINGLE (Modeling INterpersonal Group-Level Engagement), a modular three-stage pipeline that integrates: (1) off-the-shelf human detection and depth estimation, (2) VLM-based reasoning to classify pairwise social affiliation, and (3) a lightweight spatial aggregation algorithm to localize socially connected groups. To support this task and encourage future research, we present a new dataset of 100K urban street-view images annotated with bounding boxes and labels for both individuals and socially interacting groups. The annotations combine human-created labels and outputs from the MINGLE pipeline, ensuring semantic richness and broad coverage of real-world scenarios.

cs.CV↗

arXiv2024

ELSA: Evaluating Localization of Social Activities in Urban Streets using Open-Vocabulary Detection

Existing Open Vocabulary Detection (OVD) models exhibit a number of challenges. They often struggle with semantic consistency across diverse inputs, and are often sensitive to slight variations in input phrasing, leading to inconsistent performance. The calibration of their predictive confidence, especially in complex multi-label scenarios, remains suboptimal, frequently resulting in overconfident predictions that do not accurately reflect their context understanding. To understand these limitations, multi-label detection benchmarks are needed. A particularly challenging domain for such benchmarking is social activities. Due to the lack of multi-label benchmarks for social interactions, in this work we present ELSA: Evaluating Localization of Social Activities. ELSA draws on theoretical frameworks in urban sociology and design and uses in-the-wild street-level imagery, where the size of groups and the types of activities vary significantly. ELSA includes more than 900 manually annotated images with more than 4,300 multi-labeled bounding boxes for individual and group activities. We introduce a novel confidence score computation method NLSE and a novel Dynamic Box Aggregation (DBA) algorithm to assess semantic consistency in overlapping predictions. We report our results on the widely-used SOTA models Grounding DINO, Detic, OWL, and MDETR. Our evaluation protocol considers semantic stability and localization accuracy and further exposes the limitations of existing approaches.

cs.CV↗

arXiv2022

Towards Global-Scale Crowd+AI Techniques to Map and Assess Sidewalks for People with Disabilities

There is a lack of data on the location, condition, and accessibility of sidewalks across the world, which not only impacts where and how people travel but also fundamentally limits interactive mapping tools and urban analytics. In this paper, we describe initial work in semi-automatically building a sidewalk network topology from satellite imagery using hierarchical multi-scale attention models, inferring surface materials from street-level images using active learning-based semantic segmentation, and assessing sidewalk condition and accessibility features using Crowd+AI. We close with a call to create a database of labeled satellite and streetscape scenes for sidewalks and sidewalk accessibility issues along with standardized benchmarks.

cs.CV↗