Upload 5 files

app.py

@@ -0,0 +1,36 @@
+import streamlit as st
+from PIL import Image
+from io import BytesIO
+from image_captioning import generate_captions
+import requests
+import numpy as np
+
+def main():
+    st.title("Image Captioning App")
+    st.write("Upload an image or enter the URL of an image.")
+
+    # Image upload section
+    upload_option = st.radio("Select an option", ("Upload Image", "Image URL"))
+
+    if upload_option == "Upload Image":
+        uploaded_file = st.file_uploader("Choose an image", type=["jpg", "jpeg", "png"])
+        if uploaded_file is not None:
+            image = Image.open(uploaded_file)
+            st.image(image, caption="Uploaded Image", use_column_width=True)
+
+    else:
+        image_url = st.text_input("Enter the URL of an image")
+        if image_url:
+            try:
+                response = requests.get(image_url)
+                image = Image.open(BytesIO(response.content))
+                st.image(image, caption="Image from URL", use_column_width=True)
+            except:
+                st.error("Invalid image URL. Please enter a valid URL.")
+
+    if uploaded_file or image_url:
+        captions = generate_captions(image)
+        # Display the generated captions in the web interface
+
+if __name__ == '__main__':
+    main()

image_captioning.py

@@ -0,0 +1,57 @@
+import tensorflow as tf
+import numpy as np
+from tensorflow.keras.applications.inception_v3 import preprocess_input
+from tensorflow.keras.models import Model, load_model
+from tensorflow.keras.preprocessing import image
+from tensorflow.keras.preprocessing.sequence import pad_sequences
+
+# Paths
+token_path = 'saved_models/Flickr8k.token.txt'
+train_images_path = 'saved_models/Flickr_8k.trainImages.txt'
+test_images_path = 'saved_models/Flickr_8k.testImages.txt'
+model_path = 'saved_models/Final_Image_Captioning.h5'
+
+def preprocess_image(image_path):
+    img = image.load_img(image_path, target_size=(299, 299))
+    img = image.img_to_array(img)
+    img = np.expand_dims(img, axis=0)
+    img = preprocess_input(img)
+    return img
+
+def generate_caption(image_file):
+    # Load the pre-trained model
+    model = load_model(model_path)
+
+    # Preprocess the image
+    img = preprocess_image(image_file)
+
+    # Generate the caption
+    caption = generate_caption_from_image(model, img)
+
+    return caption
+
+def generate_caption_from_image(model, img):
+    max_length = 34
+    start_token = "<start>"
+    end_token = "<end>"
+    wordtoix = np.load("wordtoix.npy", allow_pickle=True).item()
+    ixtoword = np.load("ixtoword.npy", allow_pickle=True).item()
+
+    initial_state = [np.zeros((1, 256)), np.zeros((1, 256))]
+
+    # Generate caption using greedy search
+    caption = start_token
+    for _ in range(max_length):
+        sequence = [wordtoix[word] for word in caption.split() if word in wordtoix]
+        sequence = pad_sequences([sequence], maxlen=max_length)
+        y_pred = model.predict([img, sequence] + initial_state)
+        y_pred = np.argmax(y_pred)
+        word = ixtoword[y_pred]
+        caption += " " + word
+        if word == end_token:
+            break
+
+    # Remove start and end tokens
+    caption = " ".join(caption.split()[1:-1])
+
+    return caption

ixtoword.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64e40c24c08b638ed9151153e94b1847ef8f337316bf75625fe44730c9b260fb
+size 30612

requirements.txt

@@ -0,0 +1,4 @@
+streamlit==0.85.1
+tensorflow==2.5.0
+numpy~=1.19.2
+Pillow==9.5.0

wordtoix.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0f69c0df696abb66e4713736e6e13ab506a941928b4d8a98ffe0683880805ba
+size 30612