---
license: apache-2.0
language:
- en
pipeline_tag: depth-estimation
new_version: prs-eth/marigold-depth-v1-1
pinned: true
tags:
- depth estimation
- image analysis
- computer vision
- in-the-wild
- zero-shot
---
Marigold Depth v1-0 Model Card
This is a model card for the `marigold-depth-v1-0` model for monocular depth estimation from a single image.
The model is fine-tuned from the `stable-diffusion-2` [model](https://huggingface.co/stabilityai/stable-diffusion-2) as
described in our [CVPR'2024 paper](https://arxiv.org/abs/2312.02145) titled "Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation".
- Play with the interactive [Hugging Face Spaces demo](https://huggingface.co/spaces/prs-eth/marigold): check out how the model works with example images or upload your own.
- Use it with [diffusers](https://huggingface.co/docs/diffusers/using-diffusers/marigold_usage) to compute the results with a few lines of code.
- Get to the bottom of things with our [official codebase](https://github.com/prs-eth/marigold).
## Model Details
- **Developed by:** [Bingxin Ke](http://www.kebingxin.com/), [Anton Obukhov](https://www.obukhov.ai/), [Shengyu Huang](https://shengyuh.github.io/), [Nando Metzger](https://nandometzger.github.io/), [Rodrigo Caye Daudt](https://rcdaudt.github.io/), [Konrad Schindler](https://scholar.google.com/citations?user=FZuNgqIAAAAJ).
- **Model type:** Generative latent diffusion-based affine-invariant monocular depth estimation from a single image.
- **Language:** English.
- **License:** [Apache License License Version 2.0](https://www.apache.org/licenses/LICENSE-2.0).
- **Model Description:** This model can be used to generate an estimated depth map of an input image.
- **Resolution**: Even though any resolution can be processed, the model inherits the base diffusion model's effective resolution of roughly **768** pixels.
This means that for optimal predictions, any larger input image should be resized to make the longer side 768 pixels before feeding it into the model.
- **Steps and scheduler**: This model was designed for usage with the **DDIM** scheduler and between **10 and 50** denoising steps.
It is possible to obtain good predictions with just **one** step by overriding the `"timestep_spacing": "trailing"` setting
in the [scheduler configuration file](scheduler/scheduler_config.json) or by adding `pipe.scheduler = DDIMScheduler.from_config(pipe.scheduler.config, timestep_spacing="trailing")`
after the pipeline is loaded in the code before the first usage. For compatibility reasons we kept this `v1-0` model identical to the paper setting and provided a
[newer v1-1 model](https://huggingface.co/prs-eth/marigold-depth-v1-1) with optimal settings for all possible step configurations.
- **Outputs**:
- **Affine-invariant depth map**: The predicted values are between 0 and 1, interpolating between the near and far planes of the model's choice.
- **Uncertainty map**: Produced only when multiple predictions are ensembled with ensemble size larger than 2.
- **Resources for more information:** [Project Website](https://marigoldmonodepth.github.io/), [Paper](https://arxiv.org/abs/2312.02145), [Code](https://github.com/prs-eth/marigold).
- **Cite as:**
```bibtex
@InProceedings{ke2023repurposing,
title={Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation},
author={Bingxin Ke and Anton Obukhov and Shengyu Huang and Nando Metzger and Rodrigo Caye Daudt and Konrad Schindler},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2024}
}
```