README.md

---
license: cc-by-nc-4.0
pipeline_tag: image-to-3d
base_model: depth-anything/DA3-GIANT-1.1
tags:
- arrow
- 3d-reconstruction
- point-tracking
---

<div align="center">
<h1>ARROW</h1>
<h3>Arbitrary Reconstruction and Tracking of 4D Observations in the Wild</h3>

<p>
  <a href="https://arxiv.org/abs/2610.01314"><img src="https://img.shields.io/badge/arXiv-2610.01314-b31b1b?style=flat&amp;logo=arxiv" alt="arXiv: 2610.01314" /></a>
  <a href="https://vision.rwth-aachen.de/arrow"><img src="https://img.shields.io/badge/Project-Page-e77732?style=flat" alt="Project page" /></a>
  <a href="https://github.com/ifradlin/arrow"><img src="https://img.shields.io/badge/GitHub-Code-181717?style=flat&amp;logo=github" alt="GitHub code" /></a>
</p>

<p align="center">
  <a href="https://ifradlin.github.io">Ilya&nbsp;Fradlin</a> &nbsp;&nbsp;
  <a href="https://github.com/Schmiddo">Christian&nbsp;Schmidt</a> &nbsp;&nbsp;
  <a href="https://github.com/jenspiek">Jens&nbsp;Piekenbrinck</a><br />
  <a href="https://karimknaebel.com">Karim&nbsp;Knaebel</a> &nbsp;&nbsp;
  <a href="https://gonzalomartingarcia.com/">Gonzalo&nbsp;Martin&nbsp;Garcia</a> &nbsp;&nbsp;
  <a href="https://scholar.google.com/citations?user=ZcULDB0AAAAJ">Bastian&nbsp;Leibe</a>
</p>

<p><a href="https://www.rwth-aachen.de/">RWTH Aachen University</a></p>
</div>

## Overview

**ARROW** is a feed-forward model that unifies **3D reconstruction and 3D point tracking from arbitrary image sets**. Its point-query decoder supports observations from different viewpoints and capture times, enabling reconstruction and tracking from moving-camera videos, multiple video streams, and unordered photo collections.

## Usage

The snippet below briefly illustrates point queries and task helpers. For a quick start and more details, see [GitHub](https://github.com/ifradlin/arrow).

```python
import torch
from arrow import ARROW, Queries
from arrow.queries import pixel_grid
from arrow.tasks.reconstruction import reconstruction_queries
from arrow.tasks.tracking import tracking_queries

model = ARROW.from_pretrained("ifradlin/arrow", device="cuda")
images = torch.rand(1, 3, 3, 378, 504, device="cuda")  # RGB in [0, 1]
H, W = images.shape[-2:]

with torch.inference_mode():
    # Option 1: point queries.
    queries = Queries(pixel_grid(H, W, stride=4),
                      source_frame=0, target_frame=1, reference_frame=0)
    predictions = model(images, queries)

    # Option 2: task helpers, sharing one encoding.
    encoded = model.encode(images)
    reconstruction = model.decode(encoded, reconstruction_queries(images))
    tracks = model.decode(encoded, tracking_queries(images, stride=16))
```

Weights are released under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/) for non-commercial use.

## Citation

If you find our repository useful, please consider giving it a star ⭐ and citing our paper in your work:

```bibtex
@article{fradlin2026arrow,
    title         = {{ARROW}: Arbitrary Reconstruction and Tracking of 4D Observations in the Wild},
    author        = {Fradlin, Ilya and Schmidt, Christian and Piekenbrinck, Jens and Knaebel, Karim and Martin Garcia, Gonzalo and Leibe, Bastian},
    year          = 2026,
    journal       = {arXiv preprint arXiv:2610.01314},
}
```