move referring editing next to referring segmentation

app.py

@@ -92,7 +92,7 @@ description = """<p style='text-align: center'> <a href='https://x-decoder-vl.gi
 
 article = "The Demo is Run on X-Decoder (Focal-T)."
 
-inputs = [gr.inputs.Image(type='pil'), gr.inputs.Radio(choices=["Referring Segmentation", 'Open Vocabulary Semantic Segmentation','Open Vocabulary Instance Segmentation', "Open Vocabulary Panoptic Segmentation", "Image Captioning", "Text Retrieval", "Image/Region Retrieval", "Referring Captioning (Beta)", "Referring Editing"], type="value", default="OpenVocab Semantic Segmentation", label="Task"), gr.Textbox(label="xdecoder_text"), gr.Textbox(label="inpainting_text"), gr.Textbox(label="task_description")]
+inputs = [gr.inputs.Image(type='pil'), gr.inputs.Radio(choices=["Referring Segmentation", "Referring Editing", 'Open Vocabulary Semantic Segmentation','Open Vocabulary Instance Segmentation', "Open Vocabulary Panoptic Segmentation", "Image Captioning", "Text Retrieval", "Image/Region Retrieval", "Referring Captioning (Beta)"], type="value", default="OpenVocab Semantic Segmentation", label="Task"), gr.Textbox(label="xdecoder_text"), gr.Textbox(label="inpainting_text"), gr.Textbox(label="task_description")]
 gr.Interface(
     fn=inference,
     inputs=inputs,
@@ -107,6 +107,7 @@ gr.Interface(
     ],
     examples=[
     ["./images/fruit.jpg", "Referring Segmentation", "The larger watermelon.,The front white flower.,White tea pot.,Flower bunch.,white vase.,The peach on the left.,The brown knife.,The handkerchief.", '', 'Format: s,s,s'],
+    ["./images/apples.jpg", "Referring Editing", "the green apple", 'a red apple', 'x-decoder + ldm (inference takes ~20s),  use inpainting_text "clean and empty scene" for image inpainting'],
     ["./images/animals.png", "Open Vocabulary Semantic Segmentation", "zebra,antelope,giraffe,ostrich,sky,water,grass,sand,tree", '', 'Format: x,x,x'],
     ["./images/street.jpg", "Open Vocabulary Panoptic Segmentation", "stuff:building,sky,street,tree,rock,sidewalk;thing:car,person,traffic light", '', 'Format: stuff:x,x,x;thing:y,y,y'],
     ["./images/owls.jpeg", "Open Vocabulary Instance Segmentation", "owl", '', 'Format: y,y,y'],
@@ -114,7 +115,6 @@ gr.Interface(
     ["./images/rose.webp", "Text Retrieval", "lily,rose,peoney,tulip", '', 'Format: s,s,s'],
     ["./images/region_retrieval.png", "Image/Region Retrieval", "The tangerine on the plate.", '', 'Please describe the object in a detailed way (80 images in the pool).'],
     ["./images/landscape.jpg", "Referring Captioning (Beta)", "cloud", '', 'Please fill in a noun/noun phrase. (may start with a/the)'],
-    ["./images/apples.jpg", "Referring Editing", "the green apple", 'a red apple', 'x-decoder + ldm (inference takes ~20s),  use inpainting_text "clean and empty scene" for image inpainting'],
     ],
     title=title,
     description=description,