Upload 4 files

app.py

@@ -19,7 +19,7 @@ with gr.Blocks() as demo:
     gr.Markdown(DESCRIPTION)
     with gr.Tab(label="Florence-2 Image to Flux Prompt"):
         with gr.Row():
-            input_img = gr.Image(label="Input Picture",height=320)
+            input_img = gr.Image(label="Input Picture",height=320,type="filepath")
         with gr.Row():
             submit_btn = gr.Button(value="GenPrompt")
         with gr.Row():

feifeilib/feifeiflorencebase.py

@@ -114,6 +114,11 @@ def draw_ocr_bboxes(image, prediction):
     return image
 
 def process_image(image, task_prompt = "More Detailed Caption", text_input=None, model_id='microsoft/Florence-2-base'):
+    image = Image.open(image_path).convert("RGB")
+    base_height = 512
+    h_percent = (base_height / float(image.size[1]))
+    w_size = int((float(image.size[0]) * float(h_percent)))
+    image = image.resize((w_size, base_height), Image.LANCZOS)    
     image = Image.fromarray(image)  # Convert NumPy array to PIL Image
     if task_prompt == 'Caption':
         task_prompt = '<CAPTION>'