base64 encoding (#1)

- read from base64 (3c18f7a0804a94b8cdd4b7b8b67378fd38716933)


Co-authored-by: EduardoM Magdaleno <[email protected]>

app.py

@@ -24,7 +24,8 @@ import gradio as gr
 from gradio_imageslider import ImageSlider
 
 from huggingface_hub import hf_hub_download
-
+import base64
+from io import BytesIO
 USE_TORCH_COMPILE = False
 ENABLE_CPU_OFFLOAD = os.getenv("ENABLE_CPU_OFFLOAD", "0") == "1"
 
@@ -161,17 +162,16 @@ def prepare_image(input_image, resolution, hdr):
 
 @spaces.GPU
 @timer_func
-def gradio_process_image(input_image, resolution, num_inference_steps, strength, hdr, guidance_scale):
+def gradio_process_image(input_image, input_base64, resolution, num_inference_steps, strength, hdr, guidance_scale):
     print("Starting image processing...")
     torch.cuda.empty_cache()
+    if input_base64 and len(input_base64) > 0:
+        input_image = Image.open(BytesIO(base64.b64decode(input_base64)))
     
     condition_image = prepare_image(input_image, resolution, hdr)
     
-    # prompt = "masterpiece, best quality, highres"
-    # negative_prompt = "low quality, normal quality, ugly, blurry, blur, lowres, bad anatomy, bad hands, cropped, worst quality, verybadimagenegative_v1.3, JuggernautNegative-neg"
-
-    prompt = ""
-    negative_prompt = ""
+    prompt = "masterpiece, best quality, highres"
+    negative_prompt = "low quality, normal quality, ugly, blurry, blur, lowres, bad anatomy, bad hands, cropped, worst quality, verybadimagenegative_v1.3, JuggernautNegative-neg"
     
     options = {
         "prompt": prompt,
@@ -209,6 +209,8 @@ with gr.Blocks() as demo:
     with gr.Row():
         with gr.Column():
             input_image = gr.Image(type="pil", label="Input Image")
+            input_base64 = gr.Textbox(label="OR Paste Base64 Encoded Image", lines=3)
+
             run_button = gr.Button("Enhance Image")
         with gr.Column():
             output_slider = ImageSlider(label="Before / After", type="numpy")
@@ -220,7 +222,7 @@ with gr.Blocks() as demo:
         guidance_scale = gr.Slider(minimum=0, maximum=20, value=3, step=0.5, label="Guidance Scale")
 
     run_button.click(fn=gradio_process_image, 
-                     inputs=[input_image, resolution, num_inference_steps, strength, hdr, guidance_scale],
+                     inputs=[input_image, input_base64, resolution, num_inference_steps, strength, hdr, guidance_scale],
                      outputs=output_slider)
 
     # # Add examples with all required inputs