Upload 3 files

---
license: apache-2.0
library_name: onnx
tags:
- depth-estimation
- monocular-depth-estimation
- vision
---

# Depth Anything V2 - ONNX Models

This repository contains the **ONNX export versions** of the Depth Anything V2 models. These models have been converted to ONNX format for broader compatibility and inference optimization (e.g., using ONNX Runtime).

## Model Files Description

The uploaded files correspond to different model scales:

- **`depth_anything_v2_vits.onnx`**: **Small** model (Fastest inference, lightweight)
- **`depth_anything_v2_vitb.onnx`**: **Base** model (Balanced performance and speed)
- **`depth_anything_v2_vitl.onnx`**: **Large** model (Highest accuracy)

## Attribution & Original Work

This work is based on **Depth Anything V2**, originally developed by the Depth Anything Team.

- **Original Repository**: [https://github.com/DepthAnything/Depth-Anything-V2](https://github.com/DepthAnything/Depth-Anything-V2)
- **Paper**: [Depth Anything V2](https://arxiv.org/abs/2406.09414)
- **Authors**: Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.

## License

These models are distributed under the **Apache License 2.0**, consistent with the original repository.

depth_anything_v2_vitb.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83c9029c09d03d05ee3259405a8e09d7ade7bd663f29f2b7b8c0b06f5f09e554
+size 391328137

depth_anything_v2_vitl.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8779462763a10c7fbfa9700a948a9d3664f39424e8a7d78fc924f529f72a505
+size 1341282174

depth_anything_v2_vits.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b47f3b67a5fadc18dec8cfae9954544182b79b57e8beedf6051a27cd73001c77
+size 101472019