update doc

README.md

@@ -29,27 +29,11 @@ image = pipe("a photo of an astronaut riding a horse on mars",
 
 ## Decoder Finetuning
 
-This model was finetuned from the original StableDiffusion-XL autoencoder with the additional objective of including a weak invisible watermark, following the procedure of [StableSignature](https://ai.meta.com/blog/stable-signature-watermarking-generative-ai/). This watermark is decodable by [IMATAG](https://www.imatag.com/)'s demo BZH decoder, available via API.
+This model was finetuned from the original StableDiffusion-XL autoencoder with the additional objective of including a weak invisible watermark, following the procedure of [StableSignature](https://ai.meta.com/blog/stable-signature-watermarking-generative-ai/). This watermark is detectable by [IMATAG](https://www.imatag.com/)'s demo BZH decoder, available via API.
 
 ## Detecting the watermark
 
-For security reasons we cannot release the weights of the detector. It may be accessed via API to check for the presence of a watermark like this:
-```py
-import requests
-import base64
-import sys
-
-b64 = base64.b64encode(open(sys.argv[1], 'rb').read()).decode('utf8')
-data = dict(image = b64)
-response = requests.post('https://bzh.imatag.com/bzh/api/v1.0/detect',
-                         json=data)
-if response.json()['p-value'] < 1e-3:
-    print("watermark detected")
-else:
-    print("no watermark detected")
-```
-
-Or using the [detect_api.py](https://huggingface.co/spaces/imatag/stable-signature-bzh/resolve/main/detect_api.py) script:
+For security reasons we cannot release the weights of the detector. Instead, you may check for the presence of a watermark with the [detect_api.py](https://huggingface.co/spaces/imatag/stable-signature-bzh/resolve/main/detect_api.py) script like this:
 ```shell
 python detect_api.py test.png
 ```
@@ -82,3 +66,25 @@ _Visualization of watermark impact on 512x512 images from the COCO2017 validatio
 <p align="center">
 <img src=https://huggingface.co/spaces/imatag/stable-signature-bzh/resolve/main/eval/weak/0006.png />
 </p>
+
+### Robustness
+
+The watermark is robust to most editorial changes to an image, such as cropping, changing brightness or contrast, resizing or JPEG compression:
+
+<p align="center">
+  <br>
+  <b>
+from left to right: original, watermarked, difference, 'crop 50% + brigthen 50% + jpeg 80%' attack, 'downscale 2x + crop 50% + brigthen 50% + jpeg 50%' attack</b>
+</p>
+
+<p align="center">
+<img src=https://huggingface.co/spaces/imatag/stable-signature-bzh/resolve/main/eval/weak/0000.png />
+</p>
+
+The Receiver Operating Characteristic curve shows its performance under no modification and when modified by the two scenarios illustrated above:
+<p align="center">
+<img src=https://huggingface.co/spaces/imatag/stable-signature-bzh/resolve/main/eval/weak/roc.png />
+</p>
+
+More details available in our [announcement](https://www.imatag.com/blog/unlocking-the-future-of-content-authentication-imatags-breakthrough-in-ai-generated-image-watermarking) and our lab's [blog post](https://imatag-lab.medium.com/stable-signature-meets-bzh-53ad0ba13691).
+For watermarked models with a different key, other perceptual compromises, robustness to other attacks, or faster detection, please [contact IMATAG](https://pages.imatag.com/contact-us-imatag).