Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,40 +1,93 @@
-import altair as alt
-import numpy as np
-import pandas as pd
 import streamlit as st
+import torch
+import torch.nn as nn
+import timm
+import numpy as np
+import cv2
+from PIL import Image
+import io
+
+# Hide Streamlit warnings and UI elements
+st.set_page_config(layout="wide")
+st.markdown("""
+    <style>
+        footer {visibility: hidden;}
+    </style>
+""", unsafe_allow_html=True)
+
+# === Model Definition ===
+class MobileViTSegmentation(nn.Module):
+    def __init__(self, encoder_name='mobilevit_s', pretrained=False):
+        super().__init__()
+        self.backbone = timm.create_model(encoder_name, features_only=True, pretrained=pretrained)
+        self.encoder_channels = self.backbone.feature_info.channels()
+
+        self.decoder = nn.Sequential(
+            nn.Conv2d(self.encoder_channels[-1], 128, kernel_size=3, padding=1),
+            nn.Upsample(scale_factor=2, mode='bilinear'),
+            nn.Conv2d(128, 64, kernel_size=3, padding=1),
+            nn.Upsample(scale_factor=2, mode='bilinear'),
+            nn.Conv2d(64, 32, kernel_size=3, padding=1),
+            nn.Upsample(scale_factor=2, mode='bilinear'),
+            nn.Conv2d(32, 1, kernel_size=1),
+            nn.Sigmoid()
+        )
+
+    def forward(self, x):
+        feats = self.backbone(x)
+        out = self.decoder(feats[-1])
+        out = nn.functional.interpolate(out, size=(x.shape[2], x.shape[3]), mode='bilinear', align_corners=False)
+        return out
+
+# === Load Model ===
+@st.cache_resource
+def load_model():
+    model = MobileViTSegmentation()
+    state_dict = torch.load("model/mobilevit_teeth_segmentation.pth", map_location="cpu")
+    model.load_state_dict(state_dict)
+    model.eval()
+    return model
+
+model = load_model()
+
+# === Preprocessing ===
+def preprocess_image(image: Image.Image):
+    image = image.convert("RGB").resize((256, 256))
+    arr = np.array(image).astype(np.float32) / 255.0
+    arr = np.transpose(arr, (2, 0, 1))  # HWC → CHW
+    tensor = torch.tensor(arr).unsqueeze(0)  # Add batch dim
+    return tensor
+
+# === Postprocessing: Overlay Mask ===
+def overlay_mask(image_pil, mask_tensor, threshold=0.7):
+    image = np.array(image_pil.resize((256, 256)))
+    mask = mask_tensor.squeeze().detach().numpy()
+    mask_bin = (mask > threshold).astype(np.uint8) * 255
+
+    mask_color = np.zeros_like(image)
+    mask_color[..., 2] = mask_bin  # Blue mask
+
+    overlayed = cv2.addWeighted(image, 1.0, mask_color, 0.5, 0)
+    return overlayed
+
+# === UI ===
+st.title("🦷 Tooth Segmentation with MobileViT")
+st.write("Upload an image to segment the **visible teeth area** using a lightweight MobileViT segmentation model.")
+
+uploaded_file = st.file_uploader("Upload an Image", type=["jpg", "jpeg", "png"])
+
+if uploaded_file:
+    image = Image.open(uploaded_file)
+    tensor = preprocess_image(image)
+
+    with st.spinner("Segmenting..."):
+        with torch.no_grad():
+            pred = model(tensor)[0]
+
+    overlayed_img = overlay_mask(image, pred)
 
-"""
-# Welcome to Streamlit!
-
-Edit `/streamlit_app.py` to customize this app to your heart's desire :heart:.
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
-
-In the meantime, below is an example of what you can do with just a few lines of code:
-"""
-
-num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
-num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
-
-indices = np.linspace(0, 1, num_points)
-theta = 2 * np.pi * num_turns * indices
-radius = indices
-
-x = radius * np.cos(theta)
-y = radius * np.sin(theta)
-
-df = pd.DataFrame({
-    "x": x,
-    "y": y,
-    "idx": indices,
-    "rand": np.random.randn(num_points),
-})
-
-st.altair_chart(alt.Chart(df, height=700, width=700)
-    .mark_point(filled=True)
-    .encode(
-        x=alt.X("x", axis=None),
-        y=alt.Y("y", axis=None),
-        color=alt.Color("idx", legend=None, scale=alt.Scale()),
-        size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
-    ))
+    col1, col2 = st.columns(2)
+    with col1:
+        st.image(image, caption="Original Image", use_container_width=True)
+    with col2:
+        st.image(overlayed_img, caption="Tooth Mask Overlay", use_container_width=True)