用曲面片实现可解释的3D实例分割，精度更高且无体素化伪影。
SurfDist: Interpretable Three-Dimensional Instance Segmentation Using Curved Surface Patches
- 以双三次贝塞尔三角形为基本单元，构建闭合曲面表示实例
- 在生物医学数据上优于StarDist-3D，参数更紧凑，支持任意高分辨率上采样
- 适合需要可解释表面模型的3D图像分析场景
我们提出SurfDist，一种用于三维体数据实例分割的卷积神经网络架构。SurfDist能够预测由光滑参数化曲面片（具体为双三次贝塞尔三角形）构成的封闭表面表示的实例。该方法是对流行模型StarDist-3D的改进，打破了其实例参数维度与体素分辨率的耦合关系，生成的结果可在不引入体素化伪影的前提下任意上采样至更高分辨率。对于常见于生物医学成像中的类球状实例数据集，SurfDist可实现比StarDist-3D更优的性能，同时具有更紧凑的实例参数化表达。文中详细介绍了SurfDist的技术实现，并展示了在一个合成数据集和一个真实世界数据集上的实验结果，证明了可解释的实例表面模型可与实例归属信息有效联合学习。
原文摘要 · Abstract (English)
We present SurfDist, a convolutional neural network architecture for three-dimensional volumetric instance segmentation. SurfDist enables prediction of instances represented as closed surfaces composed of smooth parametric surface patches, specifically bicubic Bézier triangles. SurfDist is a modification of the popular model architecture StarDist-3D which breaks StarDist-3D's coupling of instance parameterization dimension and instance voxel resolution, and it produces predictions which may be upsampled to arbitrarily high resolutions without introduction of voxelization artifacts. For datasets with blob-shaped instances, common in biomedical imaging, SurfDist can outperform StarDist-3D with more compact instance parameterizations. We detail SurfDist's technical implementation and show one synthetic and one real-world dataset for which it outperforms StarDist-3D. These results demonstrate that interpretable instance surface models can be learned effectively alongside instance membership.
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。