changed the model

app/caption_model.py

@@ -1,15 +1,9 @@
 from transformers import pipeline
 from PIL import Image
-import torch
 
-MODEL_NAME = "Salesforce/blip-image-captioning-base"
-MAX_LENGTH = 50
-device = "cuda" if torch.cuda.is_available() else "cpu"
-
-# Load model and processor only once at startup
-processor = BlipProcessor.from_pretrained(MODEL_NAME)
-model = BlipForConditionalGeneration.from_pretrained(MODEL_NAME).to(device)
-model.eval()
+# Load object detection model
+MODEL_NAME = "facebook/detr-resnet-50"
+detector = pipeline("object-detection", model=MODEL_NAME)
 
 def caption_image(image: Image.Image):
     # Validate input