Spaces:
Build error
Build error
Sujit Pal
commited on
Commit
•
fcc5cad
1
Parent(s):
63b4da4
fix: including image for thumbnail
Browse files
app.py
CHANGED
@@ -9,14 +9,17 @@ PAGES = {
|
|
9 |
"Retrieve Images given Image": dashboard_image2image,
|
10 |
"Find Feature in Image": dashboard_featurefinder,
|
11 |
}
|
12 |
-
st.sidebar.title("CLIP-RSICD")
|
13 |
|
|
|
|
|
14 |
st.sidebar.markdown("""
|
|
|
|
|
|
|
|
|
15 |
The CLIP model from OpenAI is trained in a self-supervised manner using
|
16 |
contrastive learning to project images and caption text onto a common
|
17 |
-
embedding space.
|
18 |
-
using the [RSICD dataset](https://github.com/201528014227051/RSICD_optimal).
|
19 |
-
Click here for [more information about our project](https://github.com/arampacha/CLIP-rsicd).
|
20 |
""")
|
21 |
selection = st.sidebar.radio("Go to", list(PAGES.keys()))
|
22 |
page = PAGES[selection]
|
|
|
9 |
"Retrieve Images given Image": dashboard_image2image,
|
10 |
"Find Feature in Image": dashboard_featurefinder,
|
11 |
}
|
|
|
12 |
|
13 |
+
st.sidebar.title("CLIP-RSICD")
|
14 |
+
st.sidebar.image("thumbnail.jpg")
|
15 |
st.sidebar.markdown("""
|
16 |
+
We have fine-tuned the CLIP model (see [Model card](https://huggingface.co/flax-community/clip-rsicd-v2))
|
17 |
+
using remote sensing images from the [RSICD dataset](https://github.com/201528014227051/RSICD_optimal).
|
18 |
+
Click here for [more information about our project](https://github.com/arampacha/CLIP-rsicd).
|
19 |
+
|
20 |
The CLIP model from OpenAI is trained in a self-supervised manner using
|
21 |
contrastive learning to project images and caption text onto a common
|
22 |
+
embedding space.
|
|
|
|
|
23 |
""")
|
24 |
selection = st.sidebar.radio("Go to", list(PAGES.keys()))
|
25 |
page = PAGES[selection]
|