Add model card

README.md

@@ -0,0 +1,31 @@
+---
+license: mit
+tags:
+- depth-estimation
+- video-depth
+- monocular-geometry
+- streaming
+---
+
+# DyFN
+
+Pretrained checkpoint for **Stabilizing Streaming Video Geometry via Dynamic Feature Normalization**.
+
+- **File:** `DyFN.pt`
+- **Parameters:** ~320M
+- **Base:** MoGe-ViT-L with ConvGRU temporal stabilizer
+- **Code:** [shawLyu/Streaming_DyFN](https://github.com/shawLyu/Streaming_DyFN)
+
+## Usage
+
+```python
+from moge.model.v1 import MoGeModel
+
+model = MoGeModel.from_pretrained("shawlyu/DyFN")
+```
+
+Or pass a local path:
+
+```python
+model = MoGeModel.from_pretrained("./pretrained/DyFN.pt")
+```