# Z-upscaling: Optical Flow Guided Frame Interpolation for Isotropic Reconstruction of 3D EM Volumes

arXiv: https://arxiv.org/abs/2410.07043 · HTML: https://mlsanity.com/paper/2410.07043/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 15 references that are in the map.
- First posted on arXiv 9 Oct 2024.

## Builds on (most cited first, 9 of 9)

- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [FlowNet: Learning Optical Flow with Convolutional Networks](https://mlsanity.com/paper/1504.06852/) (2015)
- [Video Enhancement with Task-Oriented Flow](https://mlsanity.com/paper/1711.09078/) (2017)
- [Depth-Aware Video Frame Interpolation](https://mlsanity.com/paper/1904.00830/) (2019)
- [FILM: Frame Interpolation for Large Motion](https://mlsanity.com/paper/2202.04901/) (2022)
- [FLAVR: Flow-Agnostic Video Representations for Fast Frame Interpolation](https://mlsanity.com/paper/2012.08512/) (2020)
- [Deep Learning for Isotropic Super-Resolution from Non-Isotropic 3D Electron Microscopy](https://mlsanity.com/paper/1706.03142/) (2017)
- [Reference-Free Isotropic 3D EM Reconstruction using Diffusion Models](https://mlsanity.com/paper/2308.01594/) (2023)
- [SSTM: Spatiotemporal Recurrent Transformers for Multi-frame Optical Flow Estimation](https://mlsanity.com/paper/2304.14418/) (2023)

## Built on by (most cited first, 2 of 2)

- [EMGauss: Continuous Slice-to-3D Reconstruction via Dynamic Gaussian Modeling in Volume Electron Microscopy](https://mlsanity.com/paper/2512.06684/) (2025)
- [VEMamba: Efficient Isotropic Reconstruction of Volume Electron Microscopy with Axial-Lateral Consistent Mamba](https://mlsanity.com/paper/2603.00887/) (2026)

## Abstract (arXiv)

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.