---
license: mit
language:
- en
pipeline_tag: image-feature-extraction
tags:
- Foot
- reconstruction
---

<p align="center">
  <a href="http://ollieboyne.github.io/FOCUS">
        <img width=70% src="https://www.ollieboyne.com/FOCUS/images/logos/focus_v1.png">
  </a>
</p>

This repository contains the model predicting dense correspondences from foot images, as shown in our paper:

  **FOCUS: Multi-View Foot Reconstruction from Synthetically Trained Dense Correspondences**  \
  3DV 2025 \
  [Oliver Boyne](https://ollieboyne.github.io) and [Roberto Cipolla](https://mi.eng.cam.ac.uk/~cipolla/) \
  [[arXiv]](https://arxiv.org/abs/2502.06367) [[project page]](https://ollieboyne.github.io/FOCUS/)