解决堆叠物体计数难题，实现复杂场景下精准计数
Counting Stacked Objects
- 分两步：先重建堆叠物体3D结构，再分析占据比例
- 在真实与合成数据上实现高精度计数，支持不规则堆叠
- 适合生物、交通、工业等需要精准计数的领域
视觉物体计数是计算机视觉的基础任务，广泛应用于生物医学细胞计数、交通与野生动物监测等领域。然而，现有方法难以处理3D物体堆叠场景中多数物体被遮挡的问题。为此，我们提出一种新型3D计数方法，将任务分解为两个互补子问题：从多视角图像估计物体堆叠的3D几何结构和占据比例。通过结合几何重建与基于深度学习的深度分析，该方法可准确计数容器内相同物体，即使堆叠不规则。我们在多样化的现实世界与大规模合成数据集上验证了3D Counting流程的有效性，并将公开发布这些数据集以促进后续研究。
原文摘要 · Abstract (English)
Visual object counting is a fundamental computer vision task underpinning numerous real-world applications, from cell counting in biomedicine to traffic and wildlife monitoring. However, existing methods struggle to handle the challenge of stacked 3D objects in which most objects are hidden by those above them. To address this important yet underexplored problem, we propose a novel 3D counting approach that decomposes the task into two complementary subproblems - estimating the 3D geometry of the object stack and the occupancy ratio from multi-view images. By combining geometric reconstruction and deep learning-based depth analysis, our method can accurately count identical objects within containers, even when they are irregularly stacked. We validate our 3D Counting pipeline on diverse real-world and large-scale synthetic datasets, which we will release publicly to facilitate further research.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。