Update app.py

app.py

@@ -1,6 +1,3 @@
-# Installation commands (run these first in Colab)
-# !pip install deepface==0.0.79 tensorflow==2.10.0 opencv-python-headless==4.7.0.72 gradio==3.50.2
-
 import gradio as gr
 import cv2
 import numpy as np
@@ -12,13 +9,6 @@ import os
 import shutil
 import pandas as pd
 
-# Google Drive integration (for Colab)
-try:
-    from google.colab import drive
-    drive.mount('/content/drive')
-except:
-    pass
-
 def verify_faces(img1, img2, threshold=0.6, model="VGG-Face"):
     temp_dir = tempfile.mkdtemp()
     try:
@@ -61,14 +51,15 @@ def find_faces(query_img, db_input, threshold=0.6, model="VGG-Face"):
         Image.fromarray(query_img).save(query_path) if isinstance(query_img, np.ndarray) else query_img.save(query_path)
 
         # Handle database input
-        if isinstance(db_input, str):
+        if isinstance(db_input, str) and os.path.isdir(db_input):
             db_path = db_input
         else:
             db_path = os.path.join(temp_dir, "db")
             os.makedirs(db_path, exist_ok=True)
-            for i, file in enumerate(db_input):
-                ext = os.path.splitext(file.name)[1]
-                shutil.copy(file.name, os.path.join(db_path, f"img_{i}{ext}"))
+            if db_input:
+                for i, file in enumerate(db_input):
+                    ext = os.path.splitext(file.name)[1]
+                    shutil.copy(file.name, os.path.join(db_path, f"img_{i}{ext}"))
 
         # Find faces
         try:
@@ -79,10 +70,12 @@ def find_faces(query_img, db_input, threshold=0.6, model="VGG-Face"):
                 distance_metric="cosine",
                 silent=True
             )
-        except:
-            return None, {"error": "No faces found in database"}
+        except Exception as e:
+            return None, {"error": f"Face detection failed: {str(e)}"}
 
         df = dfs[0] if isinstance(dfs, list) else dfs
+        if df.empty:
+            return None, {"error": "No matches found"}
         df = df[df['distance'] <= threshold].sort_values('distance')
 
         # Create visualization
@@ -99,10 +92,15 @@ def find_faces(query_img, db_input, threshold=0.6, model="VGG-Face"):
         for i in range(num_matches):
             if i >= len(df): break
             match_path = df.iloc[i]['identity']
-            match_img = cv2.cvtColor(cv2.imread(match_path), cv2.COLOR_BGR2RGB)
-            axes[i+1].imshow(match_img)
-            axes[i+1].set_title(f"Match {i+1}\n{df.iloc[i]['distance']:.4f}")
-            axes[i+1].axis('off')
+            if not os.path.exists(match_path):
+                continue
+            try:
+                match_img = cv2.cvtColor(cv2.imread(match_path), cv2.COLOR_BGR2RGB)
+                axes[i+1].imshow(match_img)
+                axes[i+1].set_title(f"Match {i+1}\n{df.iloc[i]['distance']:.4f}")
+                axes[i+1].axis('off')
+            except:
+                continue
 
         return fig, df[['identity', 'distance']].to_dict('records')
     
@@ -174,8 +172,8 @@ with gr.Blocks(title="Face Recognition Toolkit", theme=gr.themes.Soft()) as demo
 
         with gr.Tab("Find Faces"):
             query_img = gr.Image(label="Query Image", type="pil")
-            db_input = gr.Textbox("/content/drive/MyDrive/db", label="Database Path")
-            db_files = gr.File(file_count="multiple", label="Or Upload Images")
+            db_input = gr.Textbox("", label="Database Path (optional)")
+            db_files = gr.File(file_count="multiple", label="Upload Database Images")
             find_threshold = gr.Slider(0.1, 1.0, 0.6, label="Similarity Threshold")
             find_model = gr.Dropdown(["VGG-Face", "Facenet", "OpenFace"], value="VGG-Face")
             find_btn = gr.Button("Find Matches")
@@ -184,10 +182,9 @@ with gr.Blocks(title="Face Recognition Toolkit", theme=gr.themes.Soft()) as demo
             
             find_btn.click(
                 find_faces,
-                [query_img, db_input, find_threshold, find_model],
+                [query_img, db_files, find_threshold, find_model],
                 [find_output, find_json]
             )
-            db_files.change(lambda x: None, db_files, db_input)
 
         with gr.Tab("Analyze Face"):
             analyze_img = gr.Image(label="Input Image", type="pil")