adjust sample threshold value

app.py

@@ -69,8 +69,8 @@ def query_image(img, text_queries, owl_threshold):
     # return (img, owl_output), (img, dino_output)
     return (img, owl_output)
 
-english_candidate_labels = ["hat", "sunglass", "hair band", "glove", "arm sleeve", "watch", "singlet", "t-shirts", "energy gel", "half pants", "socks", "shoes"]
-korean_candidate_labels = ["모자", "썬글라스", "헤어밴드", "장갑", "팔토시", "시계", "싱글렛", "티셔츠", "에너지젤", "쇼츠바지", "양말", "신발"]
+english_candidate_labels = ["hat", "sunglass", "hair band", "glove", "arm sleeve", "watch", "singlet", "t-shirts", "energy gel", "half pants", "socks", "shoes", "ear phone"]
+korean_candidate_labels = ["모자", "썬글라스", "헤어밴드", "장갑", "팔토시", "시계", "싱글렛", "티셔츠", "에너지젤", "쇼츠바지", "양말", "신발", "이어폰"]
 english_candidate_labels_string = ",".join(english_candidate_labels)
 
 owl_threshold = gr.Slider(0, 1, value=0.16, label="OWL Threshold")
@@ -86,6 +86,6 @@ demo = gr.Interface(
     title="OWLv2 Demo",
     description="Compare two state-of-the-art zero-shot object detection models [OWLv2](https://huggingface.co/google/owlv2-base-patch16) . Simply enter an image and the objects you want to find with comma, or try one of the examples. Play with the threshold to filter out low confidence predictions in each model.",
     # examples=[["./bee.jpg", "bee, flower", 0.16, 0.12], ["./cats.png", "cat, fishnet", 0.16, 0.12]]
-    examples=[["./rs_sample1.jpg", english_candidate_labels_string, 0.16, 0.12], ["./rs_sample2.jpg", english_candidate_labels_string, 0.16, 0.12]]
+    examples=[["./rs_sample1.jpg", english_candidate_labels_string, 0.16, 0.12], ["./rs_sample2.jpg", english_candidate_labels_string, 0.13, 0.10]]
 )
 demo.launch(debug=True)