Crowd counting has drawn increasing attention across various fields. However, existing crowd counting tasks primarily focus on estimating the overall population, ignoring the behavioral and semantic information of different social groups within the crowd. In this paper, we aim to address a newly proposed research problem, namely fine-grained crowd counting, which involves identifying different categories of individuals and accurately counting them in static images. In order to fully leverage the categorical information in static crowd images, we propose a two-tier salient feature propagation module designed to sequentially extract semantic information from both the crowd and its surrounding environment. Additionally, we introduce a category difference loss to refine the feature representation by highlighting the differences between various crowd categories. Moreover, our proposed framework can adapt to a novel problem setup called few-example fine-grained crowd counting. This setup, unlike the original fine-grained crowd counting, requires only a few exemplar point annotations instead of dense annotations from predefined categories, making it applicable in a wider range of scenarios. The baseline model for this task can be established by substituting the loss function in our proposed model with a novel hybrid loss function that integrates point-oriented cross-entropy loss and category contrastive loss. Through comprehensive experiments, we present results in both the formulation and application of fine-grained crowd counting.

In this paper, we present hierarchical D-NURBS as a new shape modeling representation which generalizes powerful, physics-based D-NURBS for interactive geometric design. Our hierarchical D-NURBS can be viewed as a collection of standard D-NURBS finite elements, organized hierarchically in a tree structure and subject to continuity constraints across the shared boundaries of adjacent D-NURBS elements at different levels. The layered and composite construction of hierarchical D-NURBS affords users the effective creation of local features and their flexible, global/local control at different level of details. Within the framework of hierarchical D-NURBS, users can interactively sculpt NURBS geometry more intuitively and conveniently through both global and local toolkits. Based on the data structure of hierarchical D-NURBS, we have developed a prototype software equipped with various physics based toolkits in the form of geometric constraints and simulated forces. Our modeling system allows users to undertake the design tasks of point manipulation, normal editing, curvature control, curve fitting, and area sculpting in interactive graphics and CAD/CAM.