diff --git a/CHANGELOG.md b/CHANGELOG.md
index d6a0eba02f..398fd16f2a 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,6 +6,9 @@ Released on December, XX, 2022.
   - New Features:
     - Added YOLOv5 as an inference-only tool ([#360](https://github.com/opendr-eu/opendr/pull/360)).
     - Added Continual Transformer Encoders ([#317](https://github.com/opendr-eu/opendr/pull/317)).
+  - Bug Fixes:
+    - Fixed `BoundingBoxList`, `TrackingAnnotationList`, `BoundingBoxList3D` and `TrackingAnnotationList3D` confidence warnings ([#365](https://github.com/opendr-eu/opendr/pull/365)).
+    - Fixed undefined `image_id` and `segmentation` for COCO `BoundingBoxList` ([#365](https://github.com/opendr-eu/opendr/pull/365)).
 
 ## Version 1.1.1
 Released on June, 30th, 2022.
diff --git a/src/opendr/engine/target.py b/src/opendr/engine/target.py
index 4b44543731..652cba01a0 100644
--- a/src/opendr/engine/target.py
+++ b/src/opendr/engine/target.py
@@ -437,11 +437,13 @@ class BoundingBoxList(Target):
     """
     def __init__(
         self,
-        boxes,
+        boxes=None,
+        image_id=-1,
     ):
         super().__init__()
-        self.data = boxes
-        self.confidence = np.mean([box.confidence for box in self.data])
+        self.data = [] if boxes is None else boxes
+        self.image_id = image_id
+        self.__compute_confidence()
 
     @staticmethod
     def from_coco(boxes_coco, image_id=0):
@@ -451,6 +453,8 @@ def from_coco(boxes_coco, image_id=0):
         for i in range(count):
             if 'segmentation' in boxes_coco[i]:
                 segmentation = boxes_coco[i]['segmentation']
+            else:
+                segmentation = []
             if 'iscrowd' in boxes_coco[i]:
                 iscrowd = boxes_coco[i]['iscrowd']
             else:
@@ -481,10 +485,17 @@ def mot(self, with_confidence=True):
 
         return result
 
+    def add_box(self, box: BoundingBox):
+        self.data.append(box)
+        self.__compute_confidence()
+
     @property
     def boxes(self):
         return self.data
 
+    def __compute_confidence(self):
+        self.confidence = sum([box.confidence for box in self.data], 0) / max(1, len(self.data))
+
     def __getitem__(self, idx):
         return self.boxes[idx]
 
@@ -578,11 +589,11 @@ class TrackingAnnotationList(Target):
     """
     def __init__(
         self,
-        boxes,
+        annotations=None,
     ):
         super().__init__()
-        self.data = boxes
-        self.confidence = np.mean([box.confidence for box in self.data])
+        self.data = [] if annotations is None else annotations
+        self.__compute_confidence()
 
     @staticmethod
     def from_mot(data):
@@ -603,10 +614,17 @@ def mot(self, with_confidence=True):
     def bounding_box_list(self):
         return BoundingBoxList([box.bounding_box() for box in self.data])
 
+    def add_annotation(self, annotation: TrackingAnnotation):
+        self.data.append(annotation)
+        self.__compute_confidence()
+
     @property
     def boxes(self):
         return self.data
 
+    def __compute_confidence(self):
+        self.confidence = sum([box.confidence for box in self.data], 0) / max(1, len(self.data))
+
     def __getitem__(self, idx):
         return self.boxes[idx]
 
@@ -723,12 +741,12 @@ class BoundingBox3DList(Target):
     """
 
     def __init__(
-            self,
-            bounding_boxes_3d
+        self,
+        bounding_boxes_3d=None
     ):
         super().__init__()
-        self.data = bounding_boxes_3d
-        self.confidence = None if len(self.data) == 0 else np.mean([box.confidence for box in self.data])
+        self.data = [] if bounding_boxes_3d is None else bounding_boxes_3d
+        self.__compute_confidence()
 
     @staticmethod
     def from_kitti(boxes_kitti):
@@ -803,10 +821,17 @@ def kitti(self):
 
         return result
 
+    def add_box(self, box: BoundingBox3D):
+        self.data.append(box)
+        self.__compute_confidence()
+
     @property
     def boxes(self):
         return self.data
 
+    def __compute_confidence(self):
+        self.confidence = sum([box.confidence for box in self.data], 0) / max(1, len(self.data))
+
     def __getitem__(self, idx):
         return self.boxes[idx]
 
@@ -910,11 +935,11 @@ class TrackingAnnotation3DList(Target):
     """
     def __init__(
         self,
-        tracking_bounding_boxes_3d
+        annotations_3d=None
     ):
         super().__init__()
-        self.data = tracking_bounding_boxes_3d
-        self.confidence = None if len(self.data) == 0 else np.mean([box.confidence for box in self.data])
+        self.data = [] if annotations_3d is None else annotations_3d
+        self.__compute_confidence()
 
     @staticmethod
     def from_kitti(boxes_kitti, ids, frames=None):
@@ -1004,9 +1029,16 @@ def kitti(self, with_tracking_info=True):
     def boxes(self):
         return self.data
 
+    def add_annotation(self, annotation: TrackingAnnotation3D):
+        self.data.append(annotation)
+        self.__compute_confidence()
+
     def bounding_box_3d_list(self):
         return BoundingBox3DList([box.bounding_box_3d() for box in self.data])
 
+    def __compute_confidence(self):
+        self.confidence = sum([box.confidence for box in self.data], 0) / max(1, len(self.data))
+
     def __getitem__(self, idx):
         return self.boxes[idx]