taskswithcode commited on
Commit
914b677
·
1 Parent(s): 89ed471

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +8 -5
app.py CHANGED
@@ -127,23 +127,26 @@ def inference(img):
127
  }
128
  )
129
  object_list_str.append(
130
- f"{predicted_label} - X:{int(x0)} Y: {int(y0)} Width {int(width)} Height: {int(height)}|"
131
  )
132
 
133
-
 
 
 
134
 
135
  return (
136
  Image.fromarray(np.uint8(out.get_image())).convert("RGB"),
137
- f"Imagine you are an intelligent image captioner.\n For each object in a scene the X, Y coordinates and width, height are given below. Use them to describe in detail each object in the scene. In the end give an overall summary of scene\n{object_list_str}"
138
  )
139
 
140
 
141
  with gr.Blocks() as demo:
142
  gr.Markdown("# Detic+ChatGPT")
143
- gr.Markdown("Use Detic to detect objects in an image and then use ChatGPT to describe the image.")
144
 
145
  #gr.HTML("<p>You can duplicating this space and use your own session token: <a style='display:inline-block' href='https://huggingface.co/spaces/yizhangliu/chatGPT?duplicate=true'><img src='https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14' alt='Duplicate Space'></a></p>")
146
- gr.HTML("<p> Instruction on how to get session token can be seen in video <a style='display:inline-block' href='https://www.youtube.com/watch?v=TdNSj_qgdFk'><font style='color:blue;weight:bold;'>here</font></a>. Add your session token by going to settings and add under secrets. </p>")
147
 
148
  with gr.Column():
149
  inp = gr.Image(label="Input Image", type="filepath")
 
127
  }
128
  )
129
  object_list_str.append(
130
+ f"{predicted_label} - X:{int(x0)} Y: {int(y0)} Width: {int(width)} Height: {int(height)}"
131
  )
132
 
133
+
134
+ output_str = "Imagine you are an intelligent image captioner. For each object in a scene the X, Y coordinates and width, height are given below. Use them to describe in detail each object in the scene. In the end give an overall summary of scene\n"
135
+ for line in object_list_str:
136
+ output_str += line + "\n"
137
 
138
  return (
139
  Image.fromarray(np.uint8(out.get_image())).convert("RGB"),
140
+ output_str
141
  )
142
 
143
 
144
  with gr.Blocks() as demo:
145
  gr.Markdown("# Detic+ChatGPT")
146
+ gr.Markdown("Use Detic to detect objects in an image and then copy/paste into ChatGPT playground to describe the image.")
147
 
148
  #gr.HTML("<p>You can duplicating this space and use your own session token: <a style='display:inline-block' href='https://huggingface.co/spaces/yizhangliu/chatGPT?duplicate=true'><img src='https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14' alt='Duplicate Space'></a></p>")
149
+ #gr.HTML("<p> Instruction on how to get session token can be seen in video <a style='display:inline-block' href='https://www.youtube.com/watch?v=TdNSj_qgdFk'><font style='color:blue;weight:bold;'>here</font></a>. Add your session token by going to settings and add under secrets. </p>")
150
 
151
  with gr.Column():
152
  inp = gr.Image(label="Input Image", type="filepath")