MSP2P: Multi-Scale Point-based Approach for Optimal Crowd Localization Through Perspective Analysis

[EN] Image-based individual localization in densely populated scenes offers practical advantages beyond mere head counting, enabling a broader range of high-level tasks in crowd analysis. Crowd image data contain drastic changes in head sizes caused by the perspective effect. This specific challenge...

Descripción completa

Detalles Bibliográficos
Autores: Redó Nieto, David, Aramburu Retegui, Mikel, García Castaño, Jorge, Sánchez Salmerón, Antonio José|||0000-0003-1896-5356
Tipo de recurso: artículo
Fecha de publicación:2025
País:España
Institución:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:inglés
OAI Identifier:oai:dnet:riunet______::964893f28cfa4f69d136078770c2ecac
Acceso en línea:https://riunet.upv.es/handle/10251/234728
Access Level:acceso abierto
Palabra clave:Crowd localization
Multi-scale
Crowd counting
Descripción
Sumario:[EN] Image-based individual localization in densely populated scenes offers practical advantages beyond mere head counting, enabling a broader range of high-level tasks in crowd analysis. Crowd image data contain drastic changes in head sizes caused by the perspective effect. This specific challenge has not been addressed in the literature, as existing localization methods do not consider multi-scale features. To alleviate this issue, we propose a novel Multi-Scale Point-to-Point Network (MSP2P) in which a set of experts are in charge of predicting head locations a at different perspective levels. However, the training procedure requires ground-truth scale information for precise one-to-one matching. For this reason, we develop a simple yet effective method that uses neighbor density information to estimate the scale associated with each head location. Extensive experiments demonstrate that our method outperforms most state-of-the-art methods on relevant counting benchmarks without compromising performance.