在计算机视觉中, 群体分析越来越受到人们的关注, 对图像中复杂人群进行分组的方法将是群体分析领域的基础技术需要. 现有的人群社交分组方法只针对固定人数的小范围场景, 不能处理真实世界中的大场景图像. 本文提出首个面向十亿像素大场景图像的基于深度学习的细粒度人群社交分组框架, 由一种图引导的全局到局部的划分策略与一个学习隐函数表示社交对交互模式的深度社交分组网络组成. 该框架可在大范围场景图像上实现准确的人群分组. 本文方法同样适用于小场景图像, 在小场景图像数据集上的实验结果表明, 本文提出的框架相比于现有方法取得了显著的性能提升.

In computer vision, more attention has been paid to group analysis, and the group detection in images becomes the key technology of human analysis on groups. The existing social grouping methods only focus on small scenes with fixed number of persons and cannot deal with large scene images in the real world. This paper proposes the first fine-grained social grouping framework for gigapixel large scene images based on deep learning, which consists of a graph-guided global to local partition strategy and a deep grouping network that learns an implicit respresentation for social pairs. The framework has achieved accurate grouping on large scene images. Our method is also applicable to small scene images, and has outperformed the existing methods.

[Code] [Paper]

Figure 1. The framework and results of fine-grained crowd social grouping for large scene images: the subimage is the Gigapixel-level image input, partially enlarged picture, grouping result for the framework.

Figure 2. Fine-grained social grouping task framework for large scenarios.

Figure 3. The structure of deep social grouping network (DSGnet).

Figure 4. The grouping results of deep social grouping network in dense crowd scenes.

Figure 5. Qualitative comparison with the state of the art methods.

Figure 6. Grouping results under different scenarios.