Fix: resolve alternative conflicts

requirements.txt

@@ -7,4 +7,4 @@ opencv-python
 grad-cam==1.4.8
 gradio
 ultralytics
-yolov8-explainer
+torchcam

yolov8.py

@@ -1,10 +1,10 @@
-from yolov8_explainer import YOLOv8Explainer
 from ultralytics import YOLO
 import cv2
 import numpy as np
 from PIL import Image
-
-# Set random colors for detection bounding boxes
+import torch
+from torchcam.methods import GradCAM
+from torchcam.utils import overlay_mask
 COLORS = np.random.uniform(0, 255, size=(80, 3))
 
 def parse_detections_yolov8(results):
@@ -37,18 +37,25 @@ def xai_yolov8(image):
     model.to('cpu')
     model.eval()
 
-    # Initialize YOLOv8 Explainer
-    explainer = YOLOv8Explainer(model)
-
     # Run YOLO detection
     results = model(image)
     boxes, colors, names = parse_detections_yolov8(results[0])
     detections_img = draw_detections(boxes, colors, names, image.copy())
 
-    # Generate Grad-CAM visualization
-    cam_image, renormalized_cam_image = explainer.visualize(image, results)
+    # Convert image to PyTorch tensor for Grad-CAM
+    image_tensor = torch.tensor(np.array(image)).permute(2, 0, 1).unsqueeze(0).float() / 255.0
+    image_tensor = image_tensor.to('cpu')
+
+    # Initialize Grad-CAM
+    grad_cam = GradCAM(model.model, target_layer='backbone')  # You can change the target_layer if needed
+
+    # Perform Grad-CAM
+    cam_map = grad_cam(image_tensor)
+
+    # Overlay Grad-CAM mask onto original image
+    cam_image = overlay_mask(image, cam_map.squeeze(0).cpu().numpy(), alpha=0.5)
 
-    # Combine original, Grad-CAM, and renormalized Grad-CAM images
-    final_image = np.hstack((image, cam_image, renormalized_cam_image))
-    caption = "Results using YOLOv8 and YOLOv8Explainer"
+    # Combine original image and Grad-CAM image
+    final_image = np.hstack((np.array(image), cam_image))
+    caption = "Results using YOLOv8 and Grad-CAM via torchcam"
     return Image.fromarray(final_image), caption