Update app.py

app.py

@@ -1,66 +1,35 @@
 import gradio as gr
-import numpy as np
 import cv2
+import numpy as np
 
-def filter_img(input_img,filter_type):
-    img = input_img
-    
-    filtered_img = np.zeros_like(img)
-    
-    if filter_type == 'Vintage':
-        sepia_filter = np.array([[0.393, 0.769, 0.189],
-                         [0.349, 0.686, 0.168],
-                         [0.272, 0.534, 0.131]])
-
-        filtered_img = np.dot(img, sepia_filter.T)
-        filtered_img = (filtered_img / filtered_img.max() * 255).astype(np.uint8)
-
-    elif filter_type == 'Gray-sharp':
-        img = cv2.cvtColor(img,cv2.COLOR_RGB2GRAY)
-        filtered_img = cv2.addWeighted(img,4, cv2.blur(img,(128,128)),-4,128)
-    elif filter_type == 'Sharp':
-        filtered_img = cv2.addWeighted(img,4, cv2.blur(img,(128,128)),-4,128)
-    elif filter_type == 'Gray':
-        filtered_img = cv2.cvtColor(img,cv2.COLOR_RGB2GRAY)
-    elif filter_type == 'B&W':
-        filtered_img = cv2.cvtColor(img,cv2.COLOR_RGB2GRAY)
-        _,filtered_img = cv2.threshold(filtered_img,127,255,cv2.THRESH_BINARY)
-    elif filter_type == 'Magenta':
-        filtered_img[:,:,0] = img[:,:,0]
-        filtered_img[:,:,2] = img[:,:,2]
-    elif filter_type == 'Cyan':
-        filtered_img[:,:,1] = img[:,:,1]
-        filtered_img[:,:,2] = img[:,:,2]
-    elif filter_type == 'Yellow':
-        filtered_img[:,:,0] = img[:,:,0]
-        filtered_img[:,:,1] = img[:,:,1]
-    elif filter_type == 'Blue':
-        filtered_img[:,:,2] = img[:,:,2]
-    elif filter_type == 'Green':
-        filtered_img[:,:,1] = img[:,:,1]
-    elif filter_type == 'Red':
-        filtered_img[:,:,0] = img[:,:,0]
-    elif filter_type == 'Cartoon':
-        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-        blurImage = cv2.medianBlur(img, 1)
-
-        edges = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY, 9, 9)
-
-        color = cv2.bilateralFilter(img, 9, 200, 200)
-
-        cartoon = cv2.bitwise_and(color, color, mask = edges)
-
-        filtered_img = cartoon
-        
-    return filtered_img
-
-demo = gr.Interface(filter_img,
-                        inputs = [ gr.Image(type='numpy'),gr.Dropdown(["Cartoon","Gray","B&W","Sharp","Gray-sharp","Vintage","Magenta","Yellow","Cyan","Blue","Green","Red"])], 
-                        outputs=[
-                            gr.Image(label="filter")
-                        ], 
-                        title = "Image Filterz", 
-                        description="Cool image filters!"
-                    )
-
-demo.launch()
+def sepia_filter(image):
+    """
+    image comes in as a NumPy array (RGB)
+    """
+    # Convert RGB (Gradio) → BGR (OpenCV)
+    image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
+
+    sepia_kernel = np.array([
+        [0.272, 0.534, 0.131],
+        [0.349, 0.686, 0.168],
+        [0.393, 0.769, 0.189]
+    ])
+
+    sepia = cv2.transform(image, sepia_kernel)
+    sepia = np.clip(sepia, 0, 255).astype(np.uint8)
+
+    # Convert back BGR → RGB for Gradio
+    sepia = cv2.cvtColor(sepia, cv2.COLOR_BGR2RGB)
+
+    return sepia
+
+app = gr.Interface(
+    fn=sepia_filter,
+    inputs=gr.Image(type="numpy", label="Input Image"),
+    outputs=gr.Image(type="numpy", label="Sepia Output"),
+    title="Simple Sepia Filter",
+    description="Upload an image and apply a sepia filter using OpenCV."
+)
+
+if __name__ == "__main__":
+    app.launch()