README.md

---
pipeline_tag: depth-estimation
license: apache-2.0
tags:
- model_hub_mixin
- pytorch_model_hub_mixin
- panorama
- 360-depth
- 360-image
- 360-depth-estimation
---

This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) 

# DA<sup>2</sup>: Depth Anything in Any Direction

[![Page](https://img.shields.io/badge/Project-Website-pink?logo=googlechrome&logoColor=white)](https://depth-any-in-any-dir.github.io/)
[![Paper](https://img.shields.io/badge/arXiv-Paper-b31b1b?logo=arxiv&logoColor=white)](http://arxiv.org/abs/2509.26618)
[![GitHub](https://img.shields.io/github/stars/EnVision-Research/DA-2?style=default&label=GitHub%20Repo%20★&logo=github)](https://github.com/EnVision-Research/DA-2)
[![HuggingFace Demo](https://img.shields.io/badge/🤗%20HuggingFace-Demo%20-yellow)](https://huggingface.co/spaces/haodongli/DA-2)

<strong>DA<sup>2</sup> predicts dense, scale-invariant distance from a single 360&deg; panorama in an end-to-end manner, with remarkable geometric fidelity and strong zero-shot generalization.</strong>

<strong>Please visit our <a href="https://depth-any-in-any-dir.github.io/" style="color: red">project page</a> for the interactive cases and more technical details.</strong>

<strong>Please visit our <a href="https://github.com/EnVision-Research/DA-2" style="color: red">Github page</a> for how to use this model.</strong>

![teaser](assets/badges/teaser.jpg)