Update README.md

README.md

@@ -6,7 +6,7 @@ language:
 
 # Model Card for Model ID
 
-<!-- Provide a quick summary of what the model is/does. -->
+This is a face recognition model, which extracts a facial feature vector from an aligned facial image.
 
 This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
 
@@ -14,10 +14,6 @@ This modelcard aims to be a base template for new models. It has been generated
 
 ### Model Description
 
-<!-- Provide a longer summary of what this model is. -->
-
-
-
 - **Developed by:** Martin Knoche
 - **Funded by [optional]:** Technical University of Munich
 - **Shared by [optional]:** Martin Knoche
@@ -80,16 +76,26 @@ SOFTWARE.
 
 ## Uses
 
-Use the model to extract a facial feature vector from an arbitrary aligned facial image.
+Use the model to extract a facial feature vector from an arbitrary aligned facial image. You can then compare that vector to other facial feature vectors to decide for same or not same person. 
 
 ### Direct Use
 
-The model can be used by within an ONNX-Runtime environment. Face must be aligned according to: 
+The model can be used by within an ONNX-Runtime environment. 
 
+```python
+model = rt.InferenceSession("FaceTransformerOctupletLoss.onnx", providers=rt.get_available_providers())
+embedding = model.run(None, {"input_image": input_image})[0][0]
+```
 
+`input_image`-Variable
+Dimensions: 112x112x3
+Channels: Should be in RGB format
+Type: float
+Values: Between 0 and 255
 
-
-[More Information Needed]
+`embedding`-Variable
+Dimension: 512
+Type: float
 
 ## Bias, Risks, and Limitations