Fast3R ViT Large 512
fair-nc
Model source
- Kind
- Other
- File size
- 2.41 GB
- Version
- vRevision a2c770b
- Updated
- Nov 4, 2026
- Downloads
- 2.4K
- Source
- Hugging Face
Source description
⚡️Fast3R - Towards 3D Reconstruction of 1000+ Images in One Forward Pass
CVPR 2025
Project Website Paper GitHub Repo Gradio Demo Hugging Face Model
Using Fast3R in Your Own Project
To use Fast3R in your own project, you can import the Fast3R class from fast3r.models.fast3r (follow instructions from the Fast3R GitHub repo to install) and use it as a regular PyTorch model.
from fast3r.models.fast3r import Fast3R
from fast3r.models.multiview_dust3r_module import MultiViewDUSt3RLitModule
# Load the model from Hugging Face
model = Fast3R.from_pretrained("jedyang97/Fast3R_ViT_Large_512")
model = model.to("cuda")
# [Optional] Create a lightweight lightning module wrapper for the model.
# This provides functions to estimate camera poses, evaluate 3D reconstruction, etc.
# See fast3r/viz/demo.py for an example.
lit_module = MultiViewDUSt3RLitModule.load_for_inference(model)
# Set model to evaluation mode
model.eval()
lit_module.eval()
Citation
@InProceedings{Yang_2025_Fast3R,
title={Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass},
author={Jianing Yang and Alexander Sax and Kevin J. Liang and Mikael Henaff and Hao Tang and Ang Cao and Joyce Chai and Franziska Meier and Matt Feiszli},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month={June},
year={2025},
}
License
The code and models are licensed under the FAIR NC Research License.
Sources
1 source
Sources
1 sourceVerified Nov 4
Download files
Model artifacts
1 artifact
Model artifacts
1 artifactmodel.safetensors
safetensors · FP32 (tensor header) · 2.41 GB · SHA-256 1357d42e7453…0427 · Hugging Face
Download fileSource excerpts
3 excerpts
Source excerpts
3 excerpts--- license: other license_name: fair-nc license_link: LICENSE tags: - image-to-3d - model_hub_mixin - pytorch_model_hub_mixin library_name: fast3r repo_url: https://github.com/facebookresearch/fast3r --- # ⚡️Fast3R - Towards 3D Reconstruction of 1000+ Images in One Forward Pass *CVPR 2025* [](https://fast3r-3d.github.io/) [](https://arxiv.org/abs/2501.13928) [](https://github.com/facebookresearch/fast3r) [](https://fast3r.ngrok.app/) [](https://huggingface.co/jedyang97/Fast3R_ViT_Large_512/) ## Using Fast3R in Your Own Project To use Fast3R in your own project, you can import the `Fast3R` class from `fast3r.models.fast3r` (follow instructions from the [Fast3R GitHub repo](https://github.com/facebookresearch/fast3r) to install) and use it as a regular PyTorch model. ```python from fast3r.models.fast3r import Fast3R from fast3r.models.multiview_dust3r_module import MultiViewDUSt3RLitModule # Load the model from Hugging Face model = Fast3R.from_pretrained("jedyang97/Fast3R_ViT_Large_512") model = model.to("cuda") # [Optional] Create a lightweight lightning module wrapper for the model. # This provides functions to estimate camera poses, evaluate 3D reconstruction, etc. # See fast3r/viz/demo.py for an example. lit_module = MultiViewDUSt3RLitModule.load_for_inference(model) # Set model to evaluation mode model.eval() lit_module.eval() ``` ## Citation ``` @InProceedings{Yang_2025_Fast3R, title={Fast3R: Towards 3D Reconstruction of 1000+ Images in One Forward Pass}, author={Jianing Yang and Alexander Sax and Kevin J. Liang and Mikael Henaff and Hao Tang and Ang Cao and Joyce Chai and Franziska Meier and Matt Feiszli}, booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month={June}, year={2025}, } ``` ## License The code and models are licensed under the [FAIR NC Research License](LICENSE).
Source context: 2412 downloads · 29 likes · Pipeline image-to-3d · Library fast3r · Repo jedyang97/Fast3R_ViT_Large_512
Source context: 2356 downloads · 28 likes · Pipeline image-to-3d · Library fast3r · Repo jedyang97/Fast3R_ViT_Large_512
