Update app.py

app.py

@@ -26,20 +26,6 @@ You may upload any images you like and try to let the model do vision tasks foll
 Some examples: You could use "Segment the dog" for segmentation, "Detect the dog" for object detection, "Estimate the depth map of this image" for depth estimation, etc.
 """  # noqa
 
-Intro_text = """
-This space showcases a demo for our paper titled "InstructCV: Towards Universal Text-to-Image Vision Generalists." We are excited to present some impressive features of our model:
-
-1. Zero-shot Capability:
-    * Our model was trained on the MS-COCO, NYUv2, Oxford-Pets, and ADE20k datasets. However, it is not limited to these datasets. You can upload any image of your choice and prompt the model to perform various vision tasks, even if they were not part of the original training set.
-
-2. Semantic Disentangling:
-    * Our model excels at handling diverse languages and instructions for different vision tasks. You can provide instructions in different languages without worrying about task confusion. The model can effectively disentangle the semantics and understand each task separately.
-
-3. Category / Data Generalization:
-    * Feel free to explore any category and experiment with images of different styles. While our model generally performs well, please note that it may not always provide optimal results for all cases. Nonetheless, we encourage you to test its capabilities across various categories and styles.
-
-"""
-
 
 example_instructions = [
                         "Please help me detect Buzz.",
@@ -154,7 +140,7 @@ def main():
             image_cfg_scale = gr.Number(value=1.5, label=f"Image weight", interactive=False)
 
 
-        gr.Markdown(Intro_text)
+        # gr.Markdown(Intro_text)
         
         load_button.click(
             fn=load_example,