Ladybird：基于对称性的深隐场三维重建的准蒙特卡罗采样
Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetry
摘要：
深度隐场回归方法是单视点图像三维重建的有效方法。然而，不同的采样模式对重建质量的影响并没有得到很好的理解。在这项工作中，我们首先研究了点集差异对网络训练的影响。基于Faradian点采样算法，我们提出了一种采样方案，理论上鼓励更好的泛化性能，并导致快速收敛的SGD为基础的优化算法。其次，基于物体的反射对称性，提出了一种特征融合方法，解决了由于自遮挡而难以利用局部图像特征的问题。我们提出的系统Ladybird能够从单个输入图像创建高质量的3D对象重建。我们在一个大规模的3D数据集（ShapeNet）上对Ladybird进行了评估，在倒角距离、地球移动器距离和联合交叉点（IoU）方面展示了极具竞争力的结果。
英文摘要：
Deep implicit field regression methods are effective for 3D reconstruction from single-view images. However, the impact of different sampling patterns on the reconstruction quality is not well-understood. In this work, we first study the effect of point set discrepancy on the network training. Based on Farthest Point Sampling algorithm, we propose a sampling scheme that theoretically encourages better generalization performance, and results in fast convergence for SGD-based optimization algorithms. Secondly, based on the reflective symmetry of an object, we propose a feature fusion method that alleviates issues due to self-occlusions which makes it difficult to utilize local image features. Our proposed system Ladybird is able to create high quality 3D object reconstructions from a single input image. We evaluate Ladybird on a large scale 3D dataset (ShapeNet) demonstrating highly competitive results in terms of Chamfer distance, Earth Mover's distance and Intersection Over Union (IoU).
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
