debugging

app.py

@@ -17,12 +17,11 @@ processor = OwlViTProcessor.from_pretrained("google/owlvit-base-patch32")
 
 
 def image_guided_detection(img, query_img, score_threshold, nms_threshold):
-    target_sizes = torch.Tensor([img.shape[:2]])
+    target_sizes = torch.Tensor([img.size[::-1]])
+    print(target_sizes)
     print(type(img), type(query_img))
-    print(img.shape)
-    print(query_img.shape)
-    print(processor)
-    print()
+    print(img.size)
+    print(query_img.size)
     inputs = processor(query_images=query_img, images=img, return_tensors="pt").to(device)
     print(inputs)
     print()
@@ -41,6 +40,7 @@ def image_guided_detection(img, query_img, score_threshold, nms_threshold):
     )
   
     boxes, scores = results[0]["boxes"], results[0]["scores"]
+    img = np.asarray(img)
 
     for box, score in zip(boxes, scores):
         box = [int(i) for i in box.tolist()]
@@ -72,7 +72,7 @@ example / query image of the target object. To use it, simply upload an image an
 
 demo = gr.Interface(
     image_guided_detection, 
-    inputs=[gr.Image(), gr.Image(), gr.Slider(0, 1, value=0.6), gr.Slider(0, 1, value=0.3)], 
+    inputs=[gr.Image(type="pil"), gr.Image(type="pil"), gr.Slider(0, 1, value=0.6), gr.Slider(0, 1, value=0.3)], 
     outputs="image",
     title="Image-Guided Object Detection with OWL-ViT",
     description=description,