Spaces:
Sleeping
Sleeping
Update src/streamlit_app.py
Browse files- src/streamlit_app.py +97 -121
src/streamlit_app.py
CHANGED
@@ -1,28 +1,25 @@
|
|
1 |
import os
|
2 |
-
import
|
3 |
-
|
4 |
-
# Set cache directory to temp or app folder
|
5 |
-
cache_dir = os.path.join(tempfile.gettempdir(), "hf_cache")
|
6 |
-
os.makedirs(cache_dir, exist_ok=True)
|
7 |
-
|
8 |
-
os.environ["XDG_CACHE_HOME"] = cache_dir
|
9 |
-
os.environ["HF_HOME"] = cache_dir
|
10 |
-
|
11 |
-
# Now import OpenCLIPEmbeddingFunction
|
12 |
-
from chromadb.utils.embedding_functions import OpenCLIPEmbeddingFunction
|
13 |
-
|
14 |
import fitz
|
15 |
import tempfile
|
16 |
import streamlit as st
|
17 |
from PIL import Image
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
from chromadb import PersistentClient
|
19 |
from chromadb.utils.data_loaders import ImageLoader
|
20 |
from chromadb.utils.embedding_functions import OpenCLIPEmbeddingFunction
|
21 |
-
from skimage import data as skdata
|
22 |
-
from skimage.io import imsave
|
23 |
-
import uuid
|
24 |
|
25 |
-
#
|
26 |
TEMP_DIR = tempfile.gettempdir()
|
27 |
IMAGES_DIR = os.path.join(TEMP_DIR, "extracted_images")
|
28 |
DB_PATH = os.path.join(TEMP_DIR, "image_vdb")
|
@@ -40,122 +37,101 @@ def get_chroma_collection():
|
|
40 |
|
41 |
image_collection = get_chroma_collection()
|
42 |
|
43 |
-
#
|
44 |
def extract_images_from_pdf(pdf_bytes):
|
45 |
pdf = fitz.open(stream=pdf_bytes, filetype="pdf")
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
59 |
-
|
60 |
-
with open(path, "wb") as f:
|
61 |
-
f.write(img_bytes)
|
62 |
-
|
63 |
-
saved_images.append(path)
|
64 |
-
|
65 |
-
return saved_images
|
66 |
-
|
67 |
-
# === Indexing ===
|
68 |
-
def index_images(image_paths):
|
69 |
-
ids = []
|
70 |
-
uris = []
|
71 |
-
for path in sorted(image_paths):
|
72 |
-
if path.lower().endswith((".png", ".jpeg", ".jpg")):
|
73 |
ids.append(str(uuid.uuid4()))
|
74 |
uris.append(path)
|
75 |
-
|
76 |
if ids:
|
77 |
image_collection.add(ids=ids, uris=uris)
|
78 |
|
79 |
-
#
|
80 |
def query_similar_images(image_file, top_k=5):
|
81 |
with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as tmp:
|
82 |
tmp.write(image_file.read())
|
83 |
-
|
84 |
-
|
85 |
-
|
86 |
-
|
87 |
-
return results['uris'][0]
|
88 |
-
finally:
|
89 |
-
os.remove(tmp_path)
|
90 |
-
|
91 |
-
# === Demo images ===
|
92 |
-
def load_skimage_demo_images():
|
93 |
-
demo_images = {
|
94 |
-
"astronaut": skdata.astronaut(),
|
95 |
-
"coffee": skdata.coffee(),
|
96 |
-
"camera": skdata.camera(),
|
97 |
-
"chelsea": skdata.chelsea(),
|
98 |
-
"rocket": skdata.rocket()
|
99 |
-
}
|
100 |
-
saved_paths = []
|
101 |
-
|
102 |
-
for name, img in demo_images.items():
|
103 |
-
path = os.path.join(IMAGES_DIR, f"{name}.png")
|
104 |
-
imsave(path, img)
|
105 |
-
saved_paths.append(path)
|
106 |
-
|
107 |
-
return saved_paths
|
108 |
-
|
109 |
-
# === Streamlit UI ===
|
110 |
-
st.title("🔍 Image Similarity Search from PDF or Custom Dataset")
|
111 |
-
|
112 |
-
source = st.radio(
|
113 |
-
"Select Image Source",
|
114 |
-
["Upload PDF", "Upload Images", "Load Demo Dataset"],
|
115 |
-
horizontal=True
|
116 |
-
)
|
117 |
-
|
118 |
-
if source == "Upload PDF":
|
119 |
-
uploaded_pdf = st.file_uploader("📤 Upload PDF", type=["pdf"])
|
120 |
-
if uploaded_pdf:
|
121 |
-
with st.spinner("Extracting images..."):
|
122 |
-
images = extract_images_from_pdf(uploaded_pdf.read())
|
123 |
-
index_images(images)
|
124 |
-
st.success(f"{len(images)} images extracted and indexed.")
|
125 |
-
st.image(images, width=150)
|
126 |
-
|
127 |
-
elif source == "Upload Images":
|
128 |
-
uploaded_imgs = st.file_uploader(
|
129 |
-
"📤 Upload one or more images", type=["jpg", "jpeg", "png"], accept_multiple_files=True
|
130 |
-
)
|
131 |
-
if uploaded_imgs:
|
132 |
-
saved_paths = []
|
133 |
-
for img in uploaded_imgs:
|
134 |
-
img_path = os.path.join(IMAGES_DIR, img.name)
|
135 |
-
with open(img_path, "wb") as f:
|
136 |
-
f.write(img.read())
|
137 |
-
saved_paths.append(img_path)
|
138 |
-
|
139 |
-
index_images(saved_paths)
|
140 |
-
st.success(f"{len(saved_paths)} images indexed.")
|
141 |
-
st.image(saved_paths, width=150)
|
142 |
-
|
143 |
-
elif source == "Load Demo Dataset":
|
144 |
-
if st.button("🔄 Load Demo Images (skimage)"):
|
145 |
-
demo_paths = load_skimage_demo_images()
|
146 |
-
index_images(demo_paths)
|
147 |
-
st.success("Demo images loaded and indexed.")
|
148 |
-
st.image(demo_paths, width=150)
|
149 |
|
150 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
151 |
|
152 |
-
st.
|
153 |
-
|
154 |
-
|
155 |
-
|
|
|
156 |
with st.spinner("Searching..."):
|
157 |
-
|
|
|
|
|
158 |
|
159 |
-
|
160 |
-
|
161 |
-
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import os
|
2 |
+
import uuid
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
import fitz
|
4 |
import tempfile
|
5 |
import streamlit as st
|
6 |
from PIL import Image
|
7 |
+
import numpy as np
|
8 |
+
from skimage.io import imsave
|
9 |
+
from torchvision.datasets import CIFAR10
|
10 |
+
import torchvision.transforms as T
|
11 |
+
|
12 |
+
# Setup cache paths
|
13 |
+
HF_CACHE = os.path.join(tempfile.gettempdir(), "hf_cache")
|
14 |
+
os.makedirs(HF_CACHE, exist_ok=True)
|
15 |
+
os.environ["XDG_CACHE_HOME"] = HF_CACHE
|
16 |
+
os.environ["HF_HOME"] = HF_CACHE
|
17 |
+
|
18 |
from chromadb import PersistentClient
|
19 |
from chromadb.utils.data_loaders import ImageLoader
|
20 |
from chromadb.utils.embedding_functions import OpenCLIPEmbeddingFunction
|
|
|
|
|
|
|
21 |
|
22 |
+
# Directories
|
23 |
TEMP_DIR = tempfile.gettempdir()
|
24 |
IMAGES_DIR = os.path.join(TEMP_DIR, "extracted_images")
|
25 |
DB_PATH = os.path.join(TEMP_DIR, "image_vdb")
|
|
|
37 |
|
38 |
image_collection = get_chroma_collection()
|
39 |
|
40 |
+
# — PDFs & Uploads —
|
41 |
def extract_images_from_pdf(pdf_bytes):
|
42 |
pdf = fitz.open(stream=pdf_bytes, filetype="pdf")
|
43 |
+
saved = []
|
44 |
+
for i in range(len(pdf)):
|
45 |
+
for img in pdf.load_page(i).get_images(full=True):
|
46 |
+
base = pdf.extract_image(img[0])
|
47 |
+
ext = base["ext"]
|
48 |
+
path = os.path.join(IMAGES_DIR, f"pdf_p{i+1}_img{img[0]}.{ext}")
|
49 |
+
with open(path,"wb") as f: f.write(base["image"])
|
50 |
+
saved.append(path)
|
51 |
+
return saved
|
52 |
+
|
53 |
+
def index_images(paths):
|
54 |
+
ids, uris = [], []
|
55 |
+
for path in sorted(paths):
|
56 |
+
if path.lower().endswith((".jpg",".jpeg",".png")):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
57 |
ids.append(str(uuid.uuid4()))
|
58 |
uris.append(path)
|
|
|
59 |
if ids:
|
60 |
image_collection.add(ids=ids, uris=uris)
|
61 |
|
62 |
+
# — Queries —
|
63 |
def query_similar_images(image_file, top_k=5):
|
64 |
with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as tmp:
|
65 |
tmp.write(image_file.read())
|
66 |
+
tmp.flush()
|
67 |
+
res = image_collection.query(query_uris=[tmp.name], n_results=top_k)
|
68 |
+
os.remove(tmp.name)
|
69 |
+
return res['uris'][0]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
70 |
|
71 |
+
def search_images_by_text(text, top_k=5):
|
72 |
+
res = image_collection.query(query_texts=[text], n_results=top_k)
|
73 |
+
return res['uris'][0]
|
74 |
+
|
75 |
+
# — Demo Dataset: CIFAR10 (500 images) —
|
76 |
+
@st.cache_resource
|
77 |
+
def load_demo_cifar10(n=500):
|
78 |
+
dataset = CIFAR10(root=TEMP_DIR, download=True, train=True)
|
79 |
+
transform = T.ToPILImage()
|
80 |
+
saved = []
|
81 |
+
for i in range(min(n, len(dataset))):
|
82 |
+
img, label = dataset[i]
|
83 |
+
if not isinstance(img, Image.Image):
|
84 |
+
img = transform(img)
|
85 |
+
path = os.path.join(IMAGES_DIR, f"cifar10_{i}_{label}.png")
|
86 |
+
img.save(path)
|
87 |
+
saved.append(path)
|
88 |
+
return saved
|
89 |
+
|
90 |
+
# — UI Starts —
|
91 |
+
st.title("🔍 Image & Text Similarity Search with 500‑Image Demo DB")
|
92 |
+
|
93 |
+
choice = st.radio("Select data source", ["Upload PDF", "Upload Images", "Load CIFAR‑10 Demo"], horizontal=True)
|
94 |
+
|
95 |
+
if choice=="Upload PDF":
|
96 |
+
pdf = st.file_uploader("📤 Upload PDF", type=["pdf"])
|
97 |
+
if pdf:
|
98 |
+
with st.spinner("Extracting..."):
|
99 |
+
imgs = extract_images_from_pdf(pdf.read()); index_images(imgs)
|
100 |
+
st.success(f"{len(imgs)} images indexed from PDF")
|
101 |
+
st.image(imgs, width=120)
|
102 |
+
|
103 |
+
elif choice=="Upload Images":
|
104 |
+
imgs = st.file_uploader("📤 Upload images", accept_multiple_files=True, type=["jpg","jpeg","png"])
|
105 |
+
if imgs:
|
106 |
+
paths=[]
|
107 |
+
for item in imgs:
|
108 |
+
p=os.path.join(IMAGES_DIR, item.name)
|
109 |
+
with open(p,"wb") as f: f.write(item.read()); paths.append(p)
|
110 |
+
index_images(paths)
|
111 |
+
st.success(f"{len(paths)} images uploaded & indexed")
|
112 |
+
st.image(paths, width=120)
|
113 |
+
|
114 |
+
elif choice=="Load CIFAR‑10 Demo":
|
115 |
+
if st.button("🔄 Load 500 CIFAR‑10 Images"):
|
116 |
+
paths=load_demo_cifar10(500); index_images(paths)
|
117 |
+
st.success("500 CIFAR‑10 demo images loaded and indexed")
|
118 |
+
st.image(paths[:20], width=100)
|
119 |
|
120 |
+
st.divider()
|
121 |
+
st.subheader("🔎 Image-Based Search")
|
122 |
+
q = st.file_uploader("Upload a query image", type=["jpg","jpeg","png"])
|
123 |
+
if q:
|
124 |
+
st.image(q, caption="Query");
|
125 |
with st.spinner("Searching..."):
|
126 |
+
out = query_similar_images(q, top_k=5)
|
127 |
+
st.subheader("Top Image Matches")
|
128 |
+
for u in out: st.image(u, width=150)
|
129 |
|
130 |
+
st.divider()
|
131 |
+
st.subheader("📝 Text-to-Image Semantic Search")
|
132 |
+
txt = st.text_input("Enter description (e.g. 'a beach'):")
|
133 |
+
if txt:
|
134 |
+
with st.spinner("Searching..."):
|
135 |
+
out = search_images_by_text(txt, top_k=5)
|
136 |
+
st.subheader("Top Semantic Matches")
|
137 |
+
for u in out: st.image(u, width=150)
|