Depth Pro 2: Sharp and Temporally Consistent Metric Monocular Depth
Focal length estimator ยท 23.5M parameters
How to use
pip install "git+https://github.com/apple-aiml-research/ml-depth-pro-2"
It loads with any size, which leaves it out of its own repository:
import torch
from depth_pro_2 import DepthPro2
# apple/DepthPro2-H, and this estimator for photos and videos without a focal length.
model = DepthPro2.from_pretrained("apple/DepthPro2-H", precision=torch.bfloat16).to("cuda")
On its own, it estimates the focal length of a photo, or one for the frames of a clip:
from functools import partial
import torch
from depth_pro_2 import load_rgb
from depth_pro_2.model import to_normalized_tensor
from depth_pro_2.network.fov import FOVEstimator
fov = FOVEstimator.from_pretrained("apple/DepthPro2-focal").to("cuda").eval()
transform = partial(to_normalized_tensor, device=torch.device("cuda"))
image, _ = load_rgb("photo.jpg")
with torch.no_grad():
f_px = fov.focal_length_px(transform(image)[None]) # pixels of the photo; frames (T, 3, H, W)
License
The weights are released under the Apple Machine Learning Research Model license, for research use only: non-commercial scientific research and academic development.
Citation
@article{DepthPro2:2026:arxiv,
title = {Depth Pro 2: Sharp and Temporally Consistent Metric Monocular Depth},
author = {Peiyun Hu and Marko Mihajlovic and Lyndon Duong and Lahav Lipson and Wei Dong and Ama\"{e}l Delaunoy and Stephan R. Richter and Vladlen Koltun},
journal = {arXiv preprint arXiv:TODO},
year = {2026},
url = {https://arxiv.org/abs/TODO},
}
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐ Ask for provider support