Update app.py
Browse files
app.py
CHANGED
@@ -1,20 +1,20 @@
|
|
1 |
-
import torch
|
2 |
-
import gradio as gr
|
3 |
-
from PIL import Image
|
4 |
-
import scipy.io.wavfile as wavfile
|
5 |
-
|
6 |
-
# Use a pipeline as a high-level helper
|
7 |
-
from transformers import pipeline
|
8 |
-
|
9 |
-
device = "cuda" if torch.cuda.is_available() else "cpu"
|
10 |
-
caption_image = pipeline("image-to-text",
|
11 |
-
model="Salesforce/blip-image-captioning-large", device=device)
|
12 |
-
def caption_my_image(pil_image):
|
13 |
-
semantics = caption_image(images=pil_image)[0]['generated_text']
|
14 |
-
return semantics
|
15 |
-
demo = gr.Interface(fn=caption_my_image,
|
16 |
-
inputs=[gr.Image(label="Select Image",type="pil")],
|
17 |
-
outputs=[gr.Textbox(label="Image Caption")],
|
18 |
-
title="
|
19 |
-
description="THIS APPLICATION WILL BE USED TO CAPTION THE IMAGE.")
|
20 |
demo.launch()
|
|
|
1 |
+
import torch
|
2 |
+
import gradio as gr
|
3 |
+
from PIL import Image
|
4 |
+
import scipy.io.wavfile as wavfile
|
5 |
+
|
6 |
+
# Use a pipeline as a high-level helper
|
7 |
+
from transformers import pipeline
|
8 |
+
|
9 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
10 |
+
caption_image = pipeline("image-to-text",
|
11 |
+
model="Salesforce/blip-image-captioning-large", device=device)
|
12 |
+
def caption_my_image(pil_image):
|
13 |
+
semantics = caption_image(images=pil_image)[0]['generated_text']
|
14 |
+
return semantics
|
15 |
+
demo = gr.Interface(fn=caption_my_image,
|
16 |
+
inputs=[gr.Image(label="Select Image",type="pil")],
|
17 |
+
outputs=[gr.Textbox(label="Image Caption")],
|
18 |
+
title="Image Captioning",
|
19 |
+
description="THIS APPLICATION WILL BE USED TO CAPTION THE IMAGE.")
|
20 |
demo.launch()
|