Update README.md
Browse files
README.md
CHANGED
|
@@ -14,6 +14,14 @@ The model checkpoint is compatible with the transformers library.
|
|
| 14 |
|
| 15 |
Depth Anything V2 was introduced in [the paper of the same name](https://arxiv.org/abs/2406.09414) by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in [this repository](https://github.com/DepthAnything/Depth-Anything-V2).
|
| 16 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
## Model description
|
| 18 |
|
| 19 |
Depth Anything V2 leverages the [DPT](https://huggingface.co/docs/transformers/model_doc/dpt) architecture with a [DINOv2](https://huggingface.co/docs/transformers/model_doc/dinov2) backbone.
|
|
|
|
| 14 |
|
| 15 |
Depth Anything V2 was introduced in [the paper of the same name](https://arxiv.org/abs/2406.09414) by Lihe Yang et al. It uses the same architecture as the original Depth Anything release but employs synthetic data and a larger capacity teacher model to achieve much finer and robust depth predictions. This fine-tuned version for metric depth estimation was first released in [this repository](https://github.com/DepthAnything/Depth-Anything-V2).
|
| 16 |
|
| 17 |
+
**Six metric depth models** of three scales for indoor and outdoor scenes, respectively, were released and are available:
|
| 18 |
+
|
| 19 |
+
| Base Model | Params | Indoor (Hypersim) | Outdoor (Virtual KITTI 2) |
|
| 20 |
+
|:-|-:|:-:|:-:|
|
| 21 |
+
| Depth-Anything-V2-Small | 24.8M | [Model Card](https://huggingface.co/depth-anything/Depth-Anything-V2-Metric-Indoor-Small-hf) | [Model Card](https://huggingface.co/depth-anything/Depth-Anything-V2-Metric-Outdoor-Small-hf) |
|
| 22 |
+
| Depth-Anything-V2-Base | 97.5M | [Model Card](https://huggingface.co/depth-anything/Depth-Anything-V2-Metric-Indoor-Base-hf) | [Model Card](https://huggingface.co/depth-anything/Depth-Anything-V2-Metric-Outdoor-Base-hf) |
|
| 23 |
+
| Depth-Anything-V2-Large | 335.3M | [Model Card](https://huggingface.co/depth-anything/Depth-Anything-V2-Metric-Indoor-Large-hf) | [Model Card](https://huggingface.co/depth-anything/Depth-Anything-V2-Metric-Outdoor-Large-hf) |
|
| 24 |
+
|
| 25 |
## Model description
|
| 26 |
|
| 27 |
Depth Anything V2 leverages the [DPT](https://huggingface.co/docs/transformers/model_doc/dpt) architecture with a [DINOv2](https://huggingface.co/docs/transformers/model_doc/dinov2) backbone.
|