用光流技术提升3D电镜图像的纵向分辨率，实现各向同性重建。
Z-upscaling: Optical Flow Guided Frame Interpolation for Isotropic Reconstruction of 3D EM Volumes
- 基于光流估计在相邻切片间推算像素运动，生成新切片。
- 在公开电镜数据集上成功实现各向同性体素重建。
- 适合生物结构三维重构、高精度显微成像研究者使用。
我们提出一种新型基于光流的方法，用于提升各向异性3D电镜（EM）体积的轴向分辨率，实现各向同性3D重建。假设对齐良好的3D生物结构具有空间连续性，我们推理出常用于视频时间分辨率增强的光流估计技术可被迁移应用。通过利用空间梯度流估计，在z方向相邻2D切片间计算像素级运动，进而插值生成新的2D切片，实现各向同性体素。我们采用最新的视频帧插值学习方法与迁移学习技术，并在公开可用的超微结构电镜数据集上验证了该方法的有效性。
原文摘要 · Abstract (English)
We propose a novel optical flow based approach to enhance the axial resolution of anisotropic 3D EM volumes to achieve isotropic 3D reconstruction. Assuming spatial continuity of 3D biological structures in well aligned EM volumes, we reasoned that optical flow estimation techniques, often applied for temporal resolution enhancement in videos, can be utilized. Pixel level motion is estimated between neighboring 2D slices along z, using spatial gradient flow estimates to interpolate and generate new 2D slices resulting in isotropic voxels. We leverage recent state-of-the-art learning methods for video frame interpolation and transfer learning techniques, and demonstrate the success of our approach on publicly available ultrastructure EM volumes.
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。