Restructure

app.py

@@ -1,9 +1,6 @@
 import gradio as gr
 import cv2
 import numpy as np
-import torch
-from torchvision import models, transforms
-from torchvision.models.detection import FasterRCNN_ResNet50_FPN_Weights
 from PIL import Image
 import mediapipe as mp
 from fer import FER  # Facial emotion recognition
@@ -12,7 +9,6 @@ from fer import FER  # Facial emotion recognition
 # Configuration
 # -----------------------------
 SKIP_RATE = 1  # For image processing, always run the analysis
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 DESIRED_SIZE = (640, 480)
 
 # -----------------------------
@@ -20,7 +16,6 @@ DESIRED_SIZE = (640, 480)
 # -----------------------------
 posture_cache = {"landmarks": None, "text": "Initializing...", "counter": 0}
 emotion_cache = {"text": "Initializing...", "counter": 0}
-objects_cache = {"boxes": None, "text": "Initializing...", "object_list_text": "", "counter": 0}
 faces_cache = {"boxes": None, "text": "Initializing...", "counter": 0}
 
 # -----------------------------
@@ -34,19 +29,9 @@ mp_drawing = mp.solutions.drawing_utils
 mp_face_detection = mp.solutions.face_detection
 face_detection = mp_face_detection.FaceDetection(min_detection_confidence=0.5)
 
-# Object Detection using Faster R-CNN
-object_detection_model = models.detection.fasterrcnn_resnet50_fpn(
-    weights=FasterRCNN_ResNet50_FPN_Weights.DEFAULT
-)
-object_detection_model.eval().to(device)
-obj_transform = transforms.Compose([transforms.ToTensor()])
-
 # Initialize the FER emotion detector (using the FER package)
 emotion_detector = FER(mtcnn=True)
 
-# Retrieve object categories from model weights metadata
-object_categories = FasterRCNN_ResNet50_FPN_Weights.DEFAULT.meta["categories"]
-
 # -----------------------------
 # Overlay Drawing Functions
 # -----------------------------
@@ -100,27 +85,6 @@ def compute_emotion_overlay(image):
         text = "No face detected"
     return text
 
-def compute_objects_overlay(image):
-    frame_bgr = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2BGR)
-    frame_bgr_small = cv2.resize(frame_bgr, DESIRED_SIZE)
-    frame_rgb_small = cv2.cvtColor(frame_bgr_small, cv2.COLOR_BGR2RGB)
-    image_pil = Image.fromarray(frame_rgb_small)
-    img_tensor = obj_transform(image_pil).to(device)
-    with torch.no_grad():
-        detections = object_detection_model([img_tensor])[0]
-    threshold = 0.8
-    boxes = []
-    object_list = []
-    for box, score, label in zip(detections["boxes"], detections["scores"], detections["labels"]):
-        if score > threshold:
-            boxes.append(tuple(box.int().cpu().numpy()))
-            label_idx = int(label)
-            label_name = object_categories[label_idx] if label_idx < len(object_categories) else "Unknown"
-            object_list.append(f"{label_name} ({score:.2f})")
-    text = f"Detected {len(boxes)} object(s)" if boxes else "No objects detected"
-    object_list_text = " | ".join(object_list) if object_list else "None"
-    return boxes, text, object_list_text
-
 def compute_faces_overlay(image):
     frame_bgr = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2BGR)
     h, w, _ = frame_bgr.shape
@@ -179,8 +143,8 @@ def compute_facemesh_overlay(image):
                 if start_idx < len(landmark_points) and end_idx < len(landmark_points):
                     pt1 = landmark_points[start_idx]
                     pt2 = landmark_points[end_idx]
-                    cv2.line(annotated, pt1, pt2, (0, 0, 255), 1)
-                    cv2.line(mask, pt1, pt2, (0, 0, 255), 1)
+                    cv2.line(annotated, pt1, pt2, (255, 0, 0), 1)
+                    cv2.line(mask, pt1, pt2, (255, 0, 0), 1)
             # Draw green dots for each landmark
             for pt in landmark_points:
                 cv2.circle(annotated, pt, 2, (0, 255, 0), -1)
@@ -194,7 +158,7 @@ def compute_facemesh_overlay(image):
 def analyze_facemesh(image):
     annotated_image, mask_image, text = compute_facemesh_overlay(image)
     return (annotated_image, mask_image, 
-            f"<div style='color: #ff6347 !important;'>Facemesh Analysis: {text}</div>")
+            f"<div style='color: #00ff00 !important;'>Facemesh Analysis: {text}</div>")
 
 # -----------------------------
 # Main Analysis Functions for Single Image
@@ -210,7 +174,7 @@ def analyze_posture_current(image):
     output = current_frame.copy()
     if posture_cache["landmarks"]:
         output = draw_posture_overlay(output, posture_cache["landmarks"])
-    return output, f"<div style='color: #ff6347 !important;'>Posture Analysis: {posture_cache['text']}</div>"
+    return output, f"<div style='color: #00ff00 !important;'>Posture Analysis: {posture_cache['text']}</div>"
 
 def analyze_emotion_current(image):
     global emotion_cache
@@ -219,22 +183,7 @@ def analyze_emotion_current(image):
     if emotion_cache["counter"] % SKIP_RATE == 0 or emotion_cache["text"] is None:
         text = compute_emotion_overlay(image)
         emotion_cache["text"] = text
-    return current_frame, f"<div style='color: #ff6347 !important;'>Emotion Analysis: {emotion_cache['text']}</div>"
-
-def analyze_objects_current(image):
-    global objects_cache
-    objects_cache["counter"] += 1
-    current_frame = np.array(image)
-    if objects_cache["counter"] % SKIP_RATE == 0 or objects_cache["boxes"] is None:
-        boxes, text, object_list_text = compute_objects_overlay(image)
-        objects_cache["boxes"] = boxes
-        objects_cache["text"] = text
-        objects_cache["object_list_text"] = object_list_text
-    output = current_frame.copy()
-    if objects_cache["boxes"]:
-        output = draw_boxes_overlay(output, objects_cache["boxes"], (255, 255, 0))
-    combined_text = f"Object Detection: {objects_cache['text']}<br>Details: {objects_cache['object_list_text']}"
-    return output, f"<div style='color: #ff6347 !important;'>{combined_text}</div>"
+    return current_frame, f"<div style='color: #00ff00 !important;'>Emotion Analysis: {emotion_cache['text']}</div>"
 
 def analyze_faces_current(image):
     global faces_cache
@@ -247,62 +196,35 @@ def analyze_faces_current(image):
     output = current_frame.copy()
     if faces_cache["boxes"]:
         output = draw_boxes_overlay(output, faces_cache["boxes"], (0, 0, 255))
-    return output, f"<div style='color: #ff6347 !important;'>Face Detection: {faces_cache['text']}</div>"
-
-def analyze_all(image):
-    current_frame = np.array(image).copy()
-    landmarks, posture_text = compute_posture_overlay(image)
-    if landmarks:
-        current_frame = draw_posture_overlay(current_frame, landmarks)
-    emotion_text = compute_emotion_overlay(image)
-    boxes_obj, objects_text, object_list_text = compute_objects_overlay(image)
-    if boxes_obj:
-        current_frame = draw_boxes_overlay(current_frame, boxes_obj, (255, 255, 0))
-    boxes_face, faces_text = compute_faces_overlay(image)
-    if boxes_face:
-        current_frame = draw_boxes_overlay(current_frame, boxes_face, (0, 0, 255))
-    combined_text = (
-        f"<b>Posture Analysis:</b> {posture_text}<br>"
-        f"<b>Emotion Analysis:</b> {emotion_text}<br>"
-        f"<b>Object Detection:</b> {objects_text}<br>"
-        f"<b>Detected Objects:</b> {object_list_text}<br>"
-        f"<b>Face Detection:</b> {faces_text}"
-    )
-    if object_list_text and object_list_text != "None":
-        description_text = f"Image Description: The scene features {object_list_text}."
-    else:
-        description_text = "Image Description: No prominent objects detected."
-    combined_text += f"<br><br><div style='border:1px solid #ff6347; padding:10px; box-shadow: 0 0 10px #ff6347;'><b>{description_text}</b></div>"
-    combined_text_html = f"<div style='color: #ff6347 !important;'>{combined_text}</div>"
-    return current_frame, combined_text_html
+    return output, f"<div style='color: #00ff00 !important;'>Face Detection: {faces_cache['text']}</div>"
 
 # -----------------------------
-# Custom CSS (Revamped High-Contrast Neon Theme)
+# Custom CSS (Revamped High-Contrast Neon Theme with Green Glows)
 # -----------------------------
 custom_css = """
 @import url('https://fonts.googleapis.com/css2?family=Orbitron:wght@400;700&display=swap');
 body {
     background-color: #121212;
     font-family: 'Orbitron', sans-serif;
-    color: #ffffff;
+    color: #00ff00;
 }
 .gradio-container {
     background: linear-gradient(135deg, #2d2d2d, #1a1a1a);
-    border: 2px solid #ff6347;
-    box-shadow: 0 0 15px #ff6347;
+    border: 2px solid #00ff00;
+    box-shadow: 0 0 15px #00ff00;
     border-radius: 10px;
     padding: 20px;
     max-width: 1200px;
     margin: auto;
 }
 .gradio-title, .gradio-description, .tab-item, .tab-item * {
-    color: #ff6347 !important;
-    text-shadow: 0 0 10px #ff6347;
+    color: #00ff00 !important;
+    text-shadow: 0 0 10px #00ff00;
 }
 input, button, .output {
-    border: 1px solid #ff6347;
-    box-shadow: 0 0 8px #ff6347;
-    color: #ffffff;
+    border: 1px solid #00ff00;
+    box-shadow: 0 0 8px #00ff00;
+    color: #00ff00;
     background-color: #1a1a1a;
 }
 """
@@ -328,15 +250,6 @@ emotion_interface = gr.Interface(
     live=False
 )
 
-objects_interface = gr.Interface(
-    fn=analyze_objects_current,
-    inputs=gr.Image(label="Upload an Image for Object Detection"),
-    outputs=[gr.Image(type="numpy", label="Annotated Output"), gr.HTML(label="Object Detection")],
-    title="Objects",
-    description="Detects objects using a pretrained Faster R-CNN.",
-    live=False
-)
-
 faces_interface = gr.Interface(
     fn=analyze_faces_current,
     inputs=gr.Image(label="Upload an Image for Face Detection"),
@@ -346,9 +259,6 @@ faces_interface = gr.Interface(
     live=False
 )
 
-# -----------------------------
-# New Facemesh Interface (Outputs annotated image and mask)
-# -----------------------------
 facemesh_interface = gr.Interface(
     fn=analyze_facemesh,
     inputs=gr.Image(label="Upload an Image for Facemesh"),
@@ -362,31 +272,18 @@ facemesh_interface = gr.Interface(
     live=False
 )
 
-all_interface = gr.Interface(
-    fn=analyze_all,
-    inputs=gr.Image(label="Upload an Image for All Inferences"),
-    outputs=[gr.Image(type="numpy", label="Annotated Output"), gr.HTML(label="Combined Analysis")],
-    title="All Inferences",
-    description="Runs posture, emotion, object, and face detection all at once.",
-    live=False
-)
-
 tabbed_interface = gr.TabbedInterface(
     interface_list=[
         posture_interface,
         emotion_interface,
-        objects_interface,
         faces_interface,
-        facemesh_interface,
-        all_interface
+        facemesh_interface
     ],
     tab_names=[
         "Posture",
         "Emotion",
-        "Objects",
         "Faces",
-        "Facemesh",
-        "All Inferences"
+        "Facemesh"
     ]
 )
 
@@ -396,7 +293,7 @@ tabbed_interface = gr.TabbedInterface(
 demo = gr.Blocks(css=custom_css)
 with demo:
     gr.Markdown("<h1 class='gradio-title'>Multi-Analysis Image App</h1>")
-    gr.Markdown("<p class='gradio-description'>Upload an image to run high-tech analysis for posture, emotions, objects, faces, and facemesh landmarks.</p>")
+    gr.Markdown("<p class='gradio-description'>Upload an image to run high-tech analysis for posture, emotions, faces, and facemesh landmarks.</p>")
     tabbed_interface.render()
 
 if __name__ == "__main__":