In the realm of 3D-computer vision applications, point cloud few-shot learning plays a critical role. However, it poses an arduous challenge due to the sparsity, irregularity, and unordered nature of the data. Current methods rely on complex local geometric extraction techniques such as convolution, graph, and attention mechanisms, along with extensive data-driven pre-training tasks. These approaches contradict the fundamental goal of few-shot learning, which is to facilitate efficient learning. To address this issue, we propose GPr-Net (Geometric Prototypical Network), a lightweight and computationally efficient geometric prototypical network that captures the intrinsic topology of point clouds and achieves superior performance. Our proposed method, IGI++ (Intrinsic Geometry Interpreter++) employs vector-based hand-crafted intrinsic geometry interpreters and Laplace vectors to extract and evaluate point cloud morphology, resulting in improved representations for FSL (Few-Shot Learning). Additionally, Laplace vectors enable the extraction of valuable features from point clouds with fewer points. To tackle the distribution drift challenge in few-shot metric learning, we leverage hyperbolic space and demonstrate that our approach handles intra and inter-class variance better than existing point cloud few-shot learning methods. Experimental results on the ModelNet40 dataset show that GPr-Net outperforms state-of-the-art methods in few-shot learning on point clouds, achieving utmost computational efficiency that is $170\times$ better than all existing works. The code is publicly available at https://github.com/TejasAnvekar/GPr-Net.
翻译:在三维计算机视觉应用中,点云小样本学习发挥着关键作用。然而,由于数据的稀疏性、不规则性和无序性,它提出了一个艰巨的挑战。当前的方法依赖于复杂的本地几何提取技术,如卷积、图形和注意机制,以及广泛的数据驱动预训练任务。这些方法与小样本学习的基本目标相矛盾,即促进有效学习。为了解决这个问题,我们提出了轻量级、计算效率高的几何原型网络 GPr-Net (Geometric Prototypical Network),它捕捉了点云的内在拓扑结构并实现了优越的性能。我们提出的方法 IGI++ (Intrinsic Geometry Interpreter++)采用基于向量的手工制作的固有几何解释器和 Laplace 向量来提取和评估点云形态,从而得到改进的小样本学习表示。此外,Laplace 向量使得可以从点云中提取更少的点的有价值的特征。为了解决小样本度量学习中的分布漂移挑战,我们利用双曲空间,并展示了我们的方法处理类内和类间方差比现有的点云小样本学习方法更好。在 ModelNet40 数据集上的实验结果表明,与所有现有的工作相比,GPr-Net 在点云小样本学习方面表现更优秀,达到了最大的计算效率,其速度是现有方法的$170 \times$。代码可从 https://github.com/TejasAnvekar/GPr-Net 获取。