import cv2
import numpy as np
import mediapipe as mp
import gradio as gr

mp_selfie_segmentation = mp.solutions.selfie_segmentation
BG_COLOR = (255, 255, 255) 

def selfie_segmentation(image):
    # Wenden Sie das Modell auf das Eingabebild an
    with mp_selfie_segmentation.SelfieSegmentation(
    model_selection=0) as selfie_segmentation:
      image = cv2.imread(image)
      image_height, image_width, _ = image.shape

      # Convert the BGR image to RGB before processing.
      results = selfie_segmentation.process(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))

      condition = np.stack((results.segmentation_mask,) * 3, axis=-1) > 0.1
      bg_image = np.zeros(image.shape, dtype=np.uint8)
      bg_image[:] = BG_COLOR
      image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
      return np.where(condition, image, bg_image)
      #im = Image.fromarray(output_image)
      #img_var /= im.save("example.png")
      #bla = "/content/example.png"
      #bla = Image.open("/content/example.png")
      #return img_var

      #pil_image = Image.fromarray(output_image)
      #byte_io = BytesIO()
      #pil_image.save(byte_io, format='PNG')
      #byte_io.seek(0)
      #return byte_io.read()


# Start a interface in gradio
iface = gr.Interface(fn=selfie_segmentation, 
                     inputs="image", 
                     outputs="image")

# starten Sie die Schnittstelle
iface.launch()