Update app.py

app.py

@@ -1,20 +1,31 @@
-import streamlit as st
-from transformers import pipeline
+from transformers import DPTImageProcessor, DPTForDepthEstimation
+import torch
+import numpy as np
 from PIL import Image
+import requests
 
-pipeline = pipeline(task="image-classification", model="julien-c/hotdog-not-hotdog")
+url = "http://images.cocodataset.org/val2017/000000039769.jpg"
+image = Image.open(requests.get(url, stream=True).raw)
 
-st.title("Hot Dog? Or Not?")
+processor = DPTImageProcessor.from_pretrained("Intel/dpt-large")
+model = DPTForDepthEstimation.from_pretrained("Intel/dpt-large")
 
-file_name = st.file_uploader("Upload a hot dog candidate image")
+# prepare image for the model
+inputs = processor(images=image, return_tensors="pt")
 
-if file_name is not None:
-    col1, col2 = st.columns(2)
+with torch.no_grad():
+    outputs = model(**inputs)
+    predicted_depth = outputs.predicted_depth
 
-    image = Image.open(file_name)
-    col1.image(image, use_column_width=True)
-    predictions = pipeline(image)
+# interpolate to original size
+prediction = torch.nn.functional.interpolate(
+    predicted_depth.unsqueeze(1),
+    size=image.size[::-1],
+    mode="bicubic",
+    align_corners=False,
+)
 
-    col2.header("Probabilities")
-    for p in predictions:
-        col2.subheader(f"{ p['label'] }: { round(p['score'] * 100, 1)}%")
+# visualize the prediction
+output = prediction.squeeze().cpu().numpy()
+formatted = (output * 255 / np.max(output)).astype("uint8")
+depth = Image.fromarray(formatted)