testing space

.gitignore

@@ -0,0 +1,34 @@
+# Python
+*.pyc
+__pycache__/
+
+# Virtual environments
+venv/
+env/
+*.env/
+
+# Jupyter Notebook
+.ipynb_checkpoints/
+
+# Model artifacts and data files
+*.h5
+*.pkl
+*.npy
+*.npz
+*.csv
+
+# Logs
+*.log
+
+# Ignore data directory if you store large datasets
+data/
+
+# Ignore model outputs
+models/
+
+# Ignore environment configuration files
+.env
+
+# Ignore system files
+.DS_Store
+Thumbs.db

requirements.txt

@@ -0,0 +1,4 @@
+torch
+streamlit
+opencv-python
+numpy

space.py

@@ -1,31 +1,54 @@
 import torch
 import streamlit as st
-# from .detector import detect_faces
-from headshot import Headshot
+import cv2
+import numpy as np
+# from headshot import Headshot
 
 device = 'cuda' if torch.cuda.is_available() else 'cpu'
 
-model = Headshot()
+# model = Headshot()
 
-# model.load_pretrain('')
+def detect_faces(frame):
+    # Implement your face detection logic here
+    # For demonstration purposes, return a dummy bounding box
+    return [(100, 100, 200, 200)]
 
 def main():
     st.title("Headshot simulator")
 
-    video_source = st.sidebar.radio("Select video source:", ("Webcam", "Upload","Demo"))
+    video_source = st.sidebar.radio("Select video source:", ("Webcam", "Upload", "Demo"))
 
     if video_source == "Webcam":
-        pass
+        cap = cv2.VideoCapture(0)
+        while True:
+            ret, frame = cap.read()
+            if not ret:
+                break
+            faces = detect_faces(frame)
+            for face in faces:
+                x, y, w, h = face
+                cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)
+            frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+            st.image(frame, channels="RGB")
 
     elif video_source == "Demo":
-        # prediction,image = sample()
+        # prediction, image = sample()
         pass
     else:
         uploaded_file = st.sidebar.file_uploader("Choose a video file", type=["mp4", "avi"])
         if uploaded_file is not None:
             file_bytes = uploaded_file.read()
-
-
+            cap = cv2.VideoCapture(uploaded_file.name)
+            while True:
+                ret, frame = cap.read()
+                if not ret:
+                    break
+                faces = detect_faces(frame)
+                for face in faces:
+                    x, y, w, h = face
+                    cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)
+                frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+                st.image(frame, channels="RGB")
 
 if __name__ == "__main__":
-    main()
+    main()