add concurrent processing for inference slicer

supervision/detection/tools/inference_slicer.py

@@ -6,6 +6,8 @@ from supervision.detection.core import Detections, validate_inference_callback
 from supervision.detection.utils import move_boxes
 from supervision.utils.image import crop_image
 
+import concurrent.futures
+
 
 def move_detections(detections: Detections, offset: np.array) -> Detections:
     """
@@ -97,15 +99,34 @@ class InferenceSlicer:
             overlap_ratio_wh=self.overlap_ratio_wh,
         )
 
-        for offset in offsets:
-            image_slice = crop_image(image=image, xyxy=offset)
-            detections = self.callback(image_slice)
-            detections = move_detections(detections=detections, offset=offset[:2])
-            detections_list.append(detections)
+        with concurrent.futures.ThreadPoolExecutor(max_workers=8) as executor:
+            futures = [
+                executor.submit(self._run_callback, image, offset) for offset in offsets
+            ]
+            for future in concurrent.futures.as_completed(futures):
+                detections_list.append(future.result())
+
         return Detections.merge(detections_list=detections_list).with_nms(
             threshold=self.iou_threshold
         )
 
+    def _run_callback(self, image, offset) -> Detections:
+        """
+        Run the provided callback on a slice of an image.
+
+        Args:
+            image (np.ndarray): The input image on which inference needs to run
+            offset (np.ndarray): An array of shape `(4,)` containing coordinates for the slice.
+
+        Returns:
+            Detections: A collection of detections for the slice.
+        """
+        image_slice = crop_image(image=image, xyxy=offset)
+        detections = self.callback(image_slice)
+        detections = move_detections(detections=detections, offset=offset[:2])
+
+        return detections
+
     @staticmethod
     def _generate_offset(
         resolution_wh: Tuple[int, int],