Adding Sample Images.

app.py

@@ -5,10 +5,13 @@ import inference_script
 import vit_gpt2
 import os
 import warnings
+
 warnings.filterwarnings('ignore')
 
 
 def process_image_and_generate_output(image, model_selection):
+    if image is None:
+        return "Please select an image", None
 
     if model_selection == ('Basic Model (Trained only for 15 epochs without any hyperparameter tuning, utilizing '
                            'inception v3)'):
@@ -36,12 +39,31 @@ def process_image_and_generate_output(image, model_selection):
     return pred_caption, audio_content
 
 
+# Define your sample images
+# sample_images = [os.path.join(os.path.dirname(__file__), 'sample_images/1.jpg'),
+#                  os.path.join(os.path.dirname(__file__), 'sample_images/2.jpg'),
+#                  os.path.join(os.path.dirname(__file__), 'sample_images/3.jpg'),
+#                  os.path.join(os.path.dirname(__file__), 'sample_images/4.jpg'), ]
+sample_images = [
+    ["sample_images/1.jpg"],
+    ["sample_images/2.jpg"],
+    ["sample_images/3.jpg"],
+    ["sample_images/4.jpg"]
+]
+
+# Create a dropdown to select sample image
+image_input = gr.Image(label="Upload Image", sources=['upload', 'webcam'])
+
+# Create a dropdown to choose the model
+model_selection_input = gr.Radio(["Basic Model (Trained only for 15 epochs without any hyperparameter "
+                                  "tuning, utilizing inception v3)",
+                                  "ViT-GPT2 (SOTA model for Image captioning)"],
+                                 label="Choose Model")
+
 iface = gr.Interface(fn=process_image_and_generate_output,
-                     inputs=["image", gr.Radio(["Basic Model (Trained only for 15 epochs without any hyperparameter "
-                                                "tuning, utilizing inception v3)", "ViT-GPT2 (SOTA model for Image "
-                                                                                   "captioning)"], label="Choose "
-                                                                                                         "Model")],
+                     inputs=[image_input, model_selection_input],
                      outputs=["text", "audio"],
+                     examples=sample_images,
                      title="Eye For Blind | Image Captioning & TTS",
                      description="To be added")