fix

app.py

@@ -20,11 +20,11 @@ def snap(image, video):
 
 demo = gr.Interface(
     snap,
-    inputs=[gr.Image(source="webcam", tool=None, type="pil"),
+    inputs=[gr.Image(source="webcam", tool=None),
             gr.Video(source="webcam")],
-    outputs=[gr.Image(label="RGB", type="pil"),
-             gr.Image(label="predicted depth", type="pil"),
-             gr.Image(label="predicted segmentation", type="pil"),
+    outputs=[gr.Image(label="RGB"),
+             gr.Image(label="predicted depth"),
+             gr.Image(label="predicted segmentation"),
              gr.Model3D(label="3D mesh reconstruction - RGB",
                         clear_color=[1.0, 1.0, 1.0, 1.0]),
              gr.Model3D(label="3D mesh reconstruction - Segmented",

inference.py

@@ -51,4 +51,4 @@ class SegmentPredictor:
         mask_annotator = sv.MaskAnnotator()
         detections = sv.Detections.from_sam(sam_result=sam_result)
         annotated_image = mask_annotator.annotate(scene=image.copy(), detections=detections)
-        return [annotated_image]
+        return annotated_image