Update app.py

app.py

@@ -1,11 +1,15 @@
 from transformers import DetrImageProcessor, DetrForObjectDetection
+from transformers import BlipProcessor, BlipForConditionalGeneration
 import torch
 from PIL import Image
 import requests
 import gradio as gr
 
-processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50")
-model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")
+box_processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50")
+box_model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")
+
+caption_processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
+caption_model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")
 
 def predict_bounding_boxes(imageurl:str):
     try:
@@ -13,15 +17,19 @@ def predict_bounding_boxes(imageurl:str):
         response.raise_for_status() 
 
         image_data = Image.open(response.raw)
-        inputs = processor(images=image_data, return_tensors="pt")
-        outputs = model(**inputs)
+        inputs = box_processor(images=image_data, return_tensors="pt")
+        outputs = box_model(**inputs)
 
         target_sizes = torch.tensor([image_data.size[::-1]])
-        results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]
+        results = box_processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]
 
-        detections = [{"score": score.item(), "label": model.config.id2label[label.item()], "box": box.tolist()} for score, label, box in zip(results["scores"], results["labels"], results["boxes"])]
+        detections = [{"score": score.item(), "label": box_model.config.id2label[label.item()], "box": box.tolist()} for score, label, box in zip(results["scores"], results["labels"], results["boxes"])]
 
-        return {"detections": detections}
+        raw_image = image_data.convert('RGB')
+        inputs = caption_processor(raw_image, return_tensors="pt")
+        out = caption_model.generate(**inputs)
+        label = caption_processor.decode(out[0], skip_special_tokens=True)
+        return {"label": label, "detections": detections}
     
     except Exception as e:
         
@@ -29,4 +37,4 @@ def predict_bounding_boxes(imageurl:str):
 
 app = gr.Interface(fn=predict_bounding_boxes, inputs="text", outputs="json")
 app.api = True
-app.launch(share=True)
+app.launch(share=True)