Research

Paper

AI LLM March 23, 2026

EgoGroups: A Benchmark For Detecting Social Groups of People in the Wild

Authors

Jeffri Murrugarra-Llerena, Pranav Chitale, Zicheng Liu, Kai Ao, Yujin Ham, Guha Balakrishnan, Paola Cascante-Bonilla

Abstract

Social group detection, or the identification of humans involved in reciprocal interpersonal interactions (e.g., family members, friends, and customers and merchants), is a crucial component of social intelligence needed for agents transacting in the world. The few existing benchmarks for social group detection are limited by low scene diversity and reliance on third-person camera sources (e.g., surveillance footage). Consequently, these benchmarks generally lack real-world evaluation on how groups form and evolve in diverse cultural contexts and unconstrained settings. To address this gap, we introduce EgoGroups, a first-person view dataset that captures social dynamics in cities around the world. EgoGroups spans 65 countries covering low, medium, and high-crowd settings under four weather/time-of-day conditions. We include dense human annotations for person and social groups, along with rich geographic and scene metadata. Using this dataset, we performed an extensive evaluation of state-of-the-art VLM/LLMs and supervised models on their group detection capabilities. We found several interesting findings, including VLMs and LLMs can outperform supervised baselines in a zero-shot setting, while crowd density and cultural regions clearly influence model performance.

Metadata

arXiv ID: 2603.22249
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-23
Fetched: 2026-03-24 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.22249v1</id>\n    <title>EgoGroups: A Benchmark For Detecting Social Groups of People in the Wild</title>\n    <updated>2026-03-23T17:43:49Z</updated>\n    <link href='https://arxiv.org/abs/2603.22249v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.22249v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Social group detection, or the identification of humans involved in reciprocal interpersonal interactions (e.g., family members, friends, and customers and merchants), is a crucial component of social intelligence needed for agents transacting in the world. The few existing benchmarks for social group detection are limited by low scene diversity and reliance on third-person camera sources (e.g., surveillance footage). Consequently, these benchmarks generally lack real-world evaluation on how groups form and evolve in diverse cultural contexts and unconstrained settings. To address this gap, we introduce EgoGroups, a first-person view dataset that captures social dynamics in cities around the world. EgoGroups spans 65 countries covering low, medium, and high-crowd settings under four weather/time-of-day conditions. We include dense human annotations for person and social groups, along with rich geographic and scene metadata. Using this dataset, we performed an extensive evaluation of state-of-the-art VLM/LLMs and supervised models on their group detection capabilities. We found several interesting findings, including VLMs and LLMs can outperform supervised baselines in a zero-shot setting, while crowd density and cultural regions clearly influence model performance.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-23T17:43:49Z</published>\n    <arxiv:comment>Project Page: https://lab-spell.github.io/EgoGroups/</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Jeffri Murrugarra-Llerena</name>\n    </author>\n    <author>\n      <name>Pranav Chitale</name>\n    </author>\n    <author>\n      <name>Zicheng Liu</name>\n    </author>\n    <author>\n      <name>Kai Ao</name>\n    </author>\n    <author>\n      <name>Yujin Ham</name>\n    </author>\n    <author>\n      <name>Guha Balakrishnan</name>\n    </author>\n    <author>\n      <name>Paola Cascante-Bonilla</name>\n    </author>\n  </entry>"
}