init

app.py

@@ -142,6 +142,11 @@ def camera_understanding(image_src, question, seed, progress=gr.Progress(track_t
 
     img_up = Image.open(saved_paths[0]).convert("RGB")
     img_lat = Image.open(saved_paths[1]).convert("RGB")
+    
+    w, h = img_up.size
+    left = max(0, w - h)
+    img_up = img_up.crop((left, 0, w, h))
+    img_lat = img_lat.crop((left, 0, w, h))
 
     return text, [img_up], [img_lat]