Upload New File

similarity.py

+"""
+Created on Sat May 4 12:00:11 2023
+@author: okanduzyel
+"""
+
+from sentence_transformers import SentenceTransformer, util
+from PIL import Image
+import glob
+import os
+
+print('Loading CLIP Model...')
+model = SentenceTransformer('clip-ViT-B-32')
+
+image_names = list(glob.glob('./images/*.jpeg'))
+
+print("Images:", len(image_names))
+encoded_image = model.encode([Image.open(filepath) for filepath in image_names], batch_size=128, convert_to_tensor=True, show_progress_bar=True)
+
+processed_images = util.paraphrase_mining_embeddings(encoded_image)
+NUM_SIMILAR_IMAGES = 30 
+
+print('Finding duplicate images...')
+duplicates = [image for image in processed_images if image[0] >= 0.999]
+
+for score, image_id1, image_id2 in duplicates[0:NUM_SIMILAR_IMAGES]:
+    print("\nScore: {:.3f}%".format(score * 100))
+    print(image_names[image_id1])
+    print(image_names[image_id2])
+
+print('Finding near duplicate images...')
+
+threshold = 0.99
+near_duplicates = [image for image in processed_images if image[0] < threshold]
+
+for score, image_id1, image_id2 in near_duplicates[0:NUM_SIMILAR_IMAGES]:
+    print("\nScore: {:.3f}%".format(score * 100))
+    print(image_names[image_id1])
+    print(image_names[image_id2])
\ No newline at end of file