docs: update README with DA3 model info and installation steps

This commit is contained in:
limbicnation
2025-11-25 01:05:25 +01:00
parent 9956ffcfe4
commit 616d8ec3da
+9
View File
@@ -13,6 +13,8 @@ A robust custom depth estimation node for ComfyUI using Depth-Anything models to
- Depth-Anything-Large
- Depth-Anything-V2-Small
- Depth-Anything-V2-Base
- Depth-Anything-V3-Small (Requires optional dependency)
- Depth-Anything-V3-Base (Requires optional dependency)
- Post-processing options:
- Gaussian blur (adjustable radius)
- Median filtering (configurable size)
@@ -46,6 +48,11 @@ A robust custom depth estimation node for ComfyUI using Depth-Anything models to
pip install -r requirements.txt
```
4. (Optional) To enable Depth Anything V3 models:
```bash
pip install git+https://github.com/ByteDance-Seed/Depth-Anything-3.git
```
4. Restart ComfyUI to load the new custom node.
> **Note**: On first use, the node will download the selected model from Hugging Face. This may take some time depending on your internet connection.
@@ -91,6 +98,8 @@ A robust custom depth estimation node for ComfyUI using Depth-Anything models to
| Depth-Anything-Small | Good | ~1.5 GB | Fast |
| Depth-Anything-V2-Base | Better | ~2.5 GB | Medium |
| Depth-Anything-Base | Better | ~2.5 GB | Medium |
| Depth-Anything-V3-Small | Excellent | ~2.0 GB | Fast |
| Depth-Anything-V3-Base | Superior | ~2.5 GB | Medium |
| Depth-Anything-Large | Best | ~4.0 GB | Slow |
## Troubleshooting Guide