{"ID":2879105,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.16970","arxiv_id":"2508.16970","title":"Local Information Matters: A Rethink of Crowd Counting","abstract":"The motivation of this paper originates from rethinking an essential characteristic of crowd counting: individuals (heads of humans) in the crowd counting task typically occupy a very small portion of the image. This characteristic has never been the focus of existing works: they typically use the same backbone as other visual tasks and pursue a large receptive field. This drives us to propose a new model design principle of crowd counting: emphasizing local modeling capability of the model. We follow the principle and design a crowd counting model named Local Information Matters Model (LIMM). The main innovation lies in two strategies: a window partitioning design that applies grid windows to the model input, and a window-wise contrastive learning design to enhance the model's ability to distinguish between local density levels. Moreover, a global attention module is applied to the end of the model to handle the occasionally occurring large-sized individuals. Extensive experiments on multiple public datasets illustrate that the proposed model shows a significant improvement in local modeling capability (8.7\\% in MAE on the JHU-Crowd++ high-density subset for example), without compromising its ability to count large-sized ones, which achieves state-of-the-art performance. Code is available at: https://github.com/tianhangpan/LIMM.","short_abstract":"The motivation of this paper originates from rethinking an essential characteristic of crowd counting: individuals (heads of humans) in the crowd counting task typically occupy a very small portion of the image. This characteristic has never been the focus of existing works: they typically use the same backbone as othe...","url_abs":"https://arxiv.org/abs/2508.16970","url_pdf":"https://arxiv.org/pdf/2508.16970v1","authors":"[\"Tianhang Pan\",\"Xiuyi Jia\"]","published":"2025-08-23T09:45:19Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false,"code_links":[{"ID":610551,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2879105,"paper_url":"https://arxiv.org/abs/2508.16970","paper_title":"Local Information Matters: A Rethink of Crowd Counting","repo_url":"https://github.com/tianhangpan/LIMM","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}