Up

app.py

@@ -1,36 +1,38 @@
-import os
-import io
-from PIL import Image
-
-from transformers import pipeline
-
-get_completion = pipeline("image-to-text",model="Salesforce/blip-image-captioning-base")
-
-def summarize(input):
-    output = get_completion(input)
-    return output[0]['generated_text']
-    
-image_url = "https://free-images.com/sm/9596/dog_animal_greyhound_983023.jpg"
-
-get_completion(image_url)
-
-import gradio as gr 
-
-def captioner(image):
-    result = get_completion(image)
-    return result[0]['generated_text']
-
-gr.close_all()
-
-christmas_dog = "https://free-images.com/sm/9596/dog_animal_greyhound_983023.jpg"
-bird = "https://free-images.com/sm/0a00/bird_exotic_bird_green.jpg"
-cow = "https://free-images.com/sm/ee7b/cow_animal_cow_head.jpg"
-
-demo = gr.Interface(fn=captioner,
-                    inputs=[gr.Image(label="Upload image", type="pil", value=christmas_dog)],
-                    outputs=[gr.Textbox(label="Caption")],
-                    title="Image Captioning with BLIP",
-                    description="Caption any image using the BLIP model",
-                    allow_flagging="never",
-                    examples=[christmas_dog, bird, cow])
-demo.launch(share=True)
+import torch 
+import re 
+import gradio as gr
+from transformers import AutoTokenizer, ViTFeatureExtractor, VisionEncoderDecoderModel 
+
+device='cpu'
+encoder_checkpoint = "nlpconnect/vit-gpt2-image-captioning"
+decoder_checkpoint = "nlpconnect/vit-gpt2-image-captioning"
+model_checkpoint = "nlpconnect/vit-gpt2-image-captioning"
+feature_extractor = ViTFeatureExtractor.from_pretrained(encoder_checkpoint)
+tokenizer = AutoTokenizer.from_pretrained(decoder_checkpoint)
+model = VisionEncoderDecoderModel.from_pretrained(model_checkpoint).to(device)
+
+def predict(image,max_length=64, num_beams=4):
+  image = image.convert('RGB')
+  image = feature_extractor(image, return_tensors="pt").pixel_values.to(device)
+  clean_text = lambda x: x.replace('<|endoftext|>','').split('\n')[0]
+  caption_ids = model.generate(image, max_length = max_length)[0]
+  caption_text = clean_text(tokenizer.decode(caption_ids))
+  return caption_text 
+
+input = gr.inputs.Image(label="Upload any Image", type = 'pil', optional=True)
+output = gr.outputs.Textbox(type="auto",label="Captions")
+examples = [f"example{i}.jpg" for i in range(1,7)]
+
+title = "Image Captioning "
+description = "Made by : shreyasdixit.tech"
+interface = gr.Interface(
+            
+        fn=predict,
+        description=description,
+        inputs = input,
+        theme="grass",
+        outputs=output,
+        examples = examples,
+        title=title,
+    )
+interface.launch(debug=True)