README.md

---
license: other
license_name: license
license_link: LICENSE
pipeline_tag: depth-estimation
---

<div align="center">
<h1>WildCross: A Cross-Modal Large Scale Benchmark for Place Recognition and Metric Depth Estimation in Natural Environments</h1>

[**Joshua Knights**](https://scholar.google.com/citations?user=RxbGr2EAAAAJ&hl=en)<sup>1,2</sup> · **Joseph Reid**<sup>1</sup> · [**Mark Cox**](https://scholar.google.com/citations?user=Bk3UD4EAAAAJ&hl=en)<sup>1</sup>
<br>
[**Kaushik Roy**](https://bit0123.github.io/)<sup>1</sup> · [**David Hall**](https://scholar.google.com/citations?user=dosODoQAAAAJ&hl=en)<sup>1</sup> · [**Peyman Moghadam**](https://scholar.google.com.au/citations?user=QAVcuWUAAAAJ&hl=en)<sup>1,2</sup>

<sup>1</sup>DATA61, CSIRO&emsp;&emsp;&emsp;<sup>2</sup>Queensland University of Technology
<br>

<a href="https://huggingface.co/papers/2603.01475"><img src='https://img.shields.io/badge/arXiv-WildCross-red' alt='Paper PDF'></a>
<a href='https://csiro-robotics.github.io/WildCross'><img src='https://img.shields.io/badge/Project_Page-WildCross-green' alt='Project Page'></a>
<a href='https://doi.org/10.25919/5fmy-yg37'><img src='https://img.shields.io/badge/Dataset_Download-WildCross-blue'></a>
</div>

This repository contains the pre-trained checkpoints for a variety of tasks on the WildCross benchmark.

![teaser](./teaser.png)

If you find this repository useful or use the WildCross dataset in your work, please cite us using the following:
```
@inproceedings{wildcross2026,
  title={{WildCross: A Cross-Modal Large Scale Benchmark for Place Recognition and Metric Depth Estimation in Natural Environments}},
  author={Joshua Knights, Joseph Reid, Kaushik Roy, David Hall, Mark Cox, Peyman Moghadam},
  booktitle={Proceedings-IEEE International Conference on Robotics and Automation},
  pages={},
  year={2026}
}
```

## Download Instructions
Our dataset can be downloaded through the [**CSIRO Data Access Portal**](https://doi.org/10.25919/5fmy-yg37).  Detailed instructions for downloading the dataset can be found in the README file provided on the data access portal page.


## Training and Benchmarking
Here we provide pre-trained checkpoints for a variety of tasks on WildCross. 

**Visual Place Recognition**
### Checkpoints
| Model      | Checkpoint Folder|
|------------|------------|
| NetVlad | [Link](https://huggingface.co/CSIRORobotics/WildCross/tree/main/VPR/NetVLAD)       |
| MixVPR | [Link](https://huggingface.co/CSIRORobotics/WildCross/tree/main/VPR/MixVPR)       |
| SALAD | [Link](https://huggingface.co/CSIRORobotics/WildCross/tree/main/VPR/SALAD)       |
| BoQ | [Link](https://huggingface.co/CSIRORobotics/WildCross/tree/main/VPR/BoQ)       |

**Cross Modal Place Recognition**
### Checkpoints
| Model      | Checkpoint Folder|
|------------|------------|
| Lip-Loc (ResNet50) | [Link](https://huggingface.co/CSIRORobotics/WildCross/tree/main/crossmodal/resnet50)       |
| Lip-Loc (Dino-v2)  | [Link](https://huggingface.co/CSIRORobotics/WildCross/tree/main/crossmodal/dinov2)       |
| Lip-Loc (Dino-v3) | [Link](https://huggingface.co/CSIRORobotics/WildCross/tree/main/crossmodal/dinov3)       |

**Metric Depth Estimation**
### Checkpoints
| Model      | Checkpoint Folder|
|------------|------------|
| DepthAnythingV2-vits | [Link](https://huggingface.co/CSIRORobotics/WildCross/resolve/main/DepthAnythingV2/finetuned/vits.pth)       |
| DepthAnythingV2-vitb | [Link](https://huggingface.co/CSIRORobotics/WildCross/resolve/main/DepthAnythingV2/finetuned/vitb.pth)       |
| DepthAnythingV2-vitl | [Link](https://huggingface.co/CSIRORobotics/WildCross/resolve/main/DepthAnythingV2/finetuned/vitl.pth)       |

For instructions on how to use these checkpoints for training or evaluation, further instructions can be found on the [WildCross GitHub repository](https://github.com/csiro-robotics/WildCross).