diff --git a/.github/workflows/tests.yaml b/.github/workflows/tests.yaml
index 3921413d..1cfc579d 100644
--- a/.github/workflows/tests.yaml
+++ b/.github/workflows/tests.yaml
@@ -60,7 +60,7 @@ jobs:
               with:
                   python-version: ${{ matrix.python-version }}
                   cache: "pip"
-            - run: pip install -e '.[dev]'
+            - run: pip install -e '.[dev, track]'
             - run: pytest --cov=src --cov-report term-missing
     doctest:
         runs-on: ubuntu-latest
@@ -74,5 +74,5 @@ jobs:
               with:
                   python-version: ${{ matrix.python-version }}
                   cache: "pip"
-            - run: pip install -e '.[dev]'
-            - run: make -C docs doctest
+            - run: pip install -e '.[dev, track]'
+            - run: make -C docs doctest
\ No newline at end of file
diff --git a/.readthedocs.yaml b/.readthedocs.yaml
index 4106f8a3..5d3f8885 100644
--- a/.readthedocs.yaml
+++ b/.readthedocs.yaml
@@ -1,16 +1,17 @@
 version: 2
 
 build:
-    os: ubuntu-22.04
-    tools:
-      python: "3.10"
+  os: ubuntu-22.04
+  tools:
+    python: "3.10"
 
 sphinx:
-    configuration: docs/source/conf.py
+  configuration: docs/source/conf.py
 
 python:
-    install:
-        - method: pip
-          path: .
-          extra_requirements:
-            - dev
+  install:
+    - method: pip
+      path: .
+      extra_requirements:
+        - dev
+        - track
diff --git a/docs/source/_static/vision/vision_gui.gif b/docs/source/_static/vision/vision_gui.gif
new file mode 100644
index 00000000..ece8763b
Binary files /dev/null and b/docs/source/_static/vision/vision_gui.gif differ
diff --git a/docs/source/_static/vision/vision_scheme.jpeg b/docs/source/_static/vision/vision_scheme.jpeg
new file mode 100644
index 00000000..45974581
Binary files /dev/null and b/docs/source/_static/vision/vision_scheme.jpeg differ
diff --git a/docs/source/_templates/class.rst b/docs/source/_templates/class.rst
index 26ee6880..7f42c44a 100644
--- a/docs/source/_templates/class.rst
+++ b/docs/source/_templates/class.rst
@@ -18,14 +18,3 @@
    {%- endfor %}
    {% endif %}
    {% endblock %}
-
-   {% block attributes %}
-   {% if attributes %}
-   .. rubric:: {{ _('Attributes') }}
-
-   .. autosummary::
-   {% for item in attributes %}
-      ~{{ name }}.{{ item }}
-   {%- endfor %}
-   {% endif %}
-   {% endblock %}
diff --git a/docs/source/index.rst b/docs/source/index.rst
index 6f7d6c0e..8728a48d 100644
--- a/docs/source/index.rst
+++ b/docs/source/index.rst
@@ -9,6 +9,7 @@
    tICA <tica>
    onion clustering <onion>
    vision <vision>
+   track <track>
    analysis <analysis>
    data processing <data_processing>
    HDF5er <hdf5er>
@@ -74,7 +75,7 @@ or with conda::
 If you want to use the :mod:`dynsight.vision` and :mod:`dynsight.track` modules
 you will need to install a series of packages. This can be done with with pip::
 
-  $ pip install ultralytics PyYAML
+  $ pip install ultralytics PyYAML pandas trackpy
 
 
 How to contribute
diff --git a/docs/source/track.rst b/docs/source/track.rst
new file mode 100644
index 00000000..2a1049f7
--- /dev/null
+++ b/docs/source/track.rst
@@ -0,0 +1,14 @@
+track
+=====
+
+This module tracks particle trajectories in `.xyz` files that lack explicit
+particle IDs, assigning consistent identities across frames.
+It is especially useful for outputs generated by the :doc:`vision` module.
+
+Functions
+---------
+
+.. toctree::
+  :maxdepth: 1
+
+  track_xyz <_autosummary/dynsight.track.track_xyz>
diff --git a/docs/source/vision.rst b/docs/source/vision.rst
index 4f08e89e..a2668721 100644
--- a/docs/source/vision.rst
+++ b/docs/source/vision.rst
@@ -1,7 +1,43 @@
 Vision
 ======
 
-Code for computer vision and trajectory extraction from experimental videos.
+Code that contains useful tools to obtain trajectories from video files.
+The extraction is possible thanks to the optimization of a Convolutional Neural
+Network (CNN) model training to
+detect specific objects in the video. Once the positions are recovered, a
+tracking algorithm (:doc:`track`) links each object with its own identity number.
+
+.. figure:: _static/vision/vision_scheme.jpeg
+   :alt: Schematic representation of a typical dynsight.vision application.
+   :align: center
+
+   Schematic representation of a typical `dynsight.vision` application.
+
+The object detection is managed by the Ultralytics_ external library. If you
+use this tool, please cite them by referring to this link_.
+
+.. _Ultralytics: https://www.ultralytics.com
+.. _link: https://docs.ultralytics.com/models/yolo11/#citations-and-acknowledgements
+
+--------------
+The Vision GUI
+--------------
+
+The `dynsight.vision` module includes an internal Graphical User Interface (GUI)
+designed to assist users in preparing the training items required for object
+detection. It only requires an *initial guess* of the objects present in
+the video, which is then used to generate a synthetic dataset to bootstrap the
+training process.
+
+.. tip::
+
+   Applying some basic color correction to the input video (especially converting
+   it to grayscale with enhanced contrast) can significantly improve the quality
+   of the initial model.
+
+.. image:: _static/vision/vision_gui.gif
+   :alt: Usage of the GUI.
+   :align: center
 
 -----
 Usage
@@ -12,3 +48,6 @@ Usage
 
   Video <_autosummary/dynsight.vision.Video.rst>
   Detect <_autosummary/dynsight.vision.Detect.rst>
+
+Once obtained the particle positions, is possible to obtain the trajectories of
+each particle by using the :doc:`track` module.
diff --git a/examples/video_detection.py b/examples/video_detection.py
index 9d5695cb..d1d78bf2 100644
--- a/examples/video_detection.py
+++ b/examples/video_detection.py
@@ -113,10 +113,25 @@ def main() -> None:
         )
     if args.predict:
         detection.predict_frames(model_path=args.detect_model)
+        xyz_trajectory = Path.cwd() / "trajectory.xyz"
         detection.compute_xyz(
             prediction_folder_path=Path("prediction"),
             output_path=Path.cwd(),
         )
+        tr_xyz_trajectory = Path.cwd() / "tracked_trajectory.xyz"
+        trj = dynsight.track.track_xyz(
+            input_xyz=xyz_trajectory,
+            output_xyz=tr_xyz_trajectory,
+        )
+        # Usage example of the trj object after tracking.
+        lens_descriptor = trj.get_lens(r_cut=5)
+        lens_descriptor.dump_to_json(output_project)
+        # Compute clustering
+        onion_lens = lens_descriptor.get_onion(delta_t=2)
+        onion_lens.plot_output(
+            file_path=args.output,
+            data_insight=lens_descriptor,
+        )
 
 
 if __name__ == "__main__":
diff --git a/pyproject.toml b/pyproject.toml
index b1949aee..d14db4ab 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -30,6 +30,7 @@ dev = [
     "furo",
     "numba",
 ]
+track = ["trackpy", "pandas"]
 
 [project.urls]
 github = "https://github.com/GMPavanLab/dynsight"
@@ -110,6 +111,8 @@ module = [
     'Pillow.*',
     'ultralytics.*',
     'opencv-python.*',
+    'pandas.*',
+    'trackpy.*',
     'cv2.*',
 ]
 ignore_missing_imports = true
diff --git a/src/dynsight/__init__.py b/src/dynsight/__init__.py
index 36bc1232..e16b196f 100644
--- a/src/dynsight/__init__.py
+++ b/src/dynsight/__init__.py
@@ -8,6 +8,7 @@
     onion,
     soap,
     tica,
+    track,
     trajectory,
     utilities,
     vision,
@@ -21,6 +22,7 @@
     "onion",
     "soap",
     "tica",
+    "track",
     "trajectory",
     "utilities",
     "vision",
diff --git a/src/dynsight/_internal/track/__init__.py b/src/dynsight/_internal/track/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/src/dynsight/_internal/track/track.py b/src/dynsight/_internal/track/track.py
new file mode 100644
index 00000000..25ed5ff0
--- /dev/null
+++ b/src/dynsight/_internal/track/track.py
@@ -0,0 +1,138 @@
+from __future__ import annotations
+
+import logging
+from pathlib import Path
+
+import pandas as pd
+import trackpy as tp
+
+from dynsight.trajectory import Trj
+
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s | %(levelname)s | %(message)s",
+)
+logger = logging.getLogger(__name__)
+
+
+def track_xyz(
+    input_xyz: Path,
+    output_xyz: Path,
+    search_range: float = 10,
+    memory: int = 1,
+    adaptive_stop: None | float = 0.95,
+    adaptive_step: None | float = 0.5,
+) -> Trj:
+    """Track particles from an `.xyz` file and write a new file with IDs.
+
+    The input `.xyz` is assumed to contain only raw 3D coordinates
+    (without atom labels/identity), and each frame begins with a line
+    indicating the number of atoms, followed by a comment line, then a list of
+    positions. Each frame in the input file must follow this structure::
+
+        <number of atoms>
+        comment line
+        <x> <y> <z>
+        <x> <y> <z>
+        ...
+        <x> <y> <z>
+
+    The output file will have the same structure, but each line will start
+    with the tracked particle ID.
+
+    Parameters:
+        input_xyz:
+            Path to the input .xyz file containing positions only.
+
+        output_xyz:
+            Path where the output .xyz file with particle IDs will be saved.
+
+        search_range:
+            The maximum allowable displacement of objects between frames for
+            them to be considered the same particle.
+
+        memory:
+            The maximum number of frames during which an object can vanish,
+            then reppear nearby, and be considered the same particle.
+
+        adaptive_stop:
+            If not `None`, when encountering a region with too many candidate
+            links (subnet), retry by progressively reducing `search_range`
+            until the subnet is solvable. If `search_range` becomes less or
+            equal than the `adaptive_stop`, give up and raise a
+            `SubnetOversizeException`.
+
+        adaptive_step:
+            Factor by which the `search_range` is multiplied to reduce it
+            during adaptive search. Effective only if `adaptive_stop` is not
+            `None`.
+    """
+    if adaptive_stop is None and adaptive_step is not None:
+        msg = "adaptive_step is set but adaptive_stop is None."
+        raise ValueError(msg)
+    if adaptive_stop is not None and adaptive_step is None:
+        msg = "adaptive_stop is set but adaptive_step is None."
+        raise ValueError(msg)
+
+    input_xyz = Path(input_xyz)
+    output_xyz = Path(output_xyz)
+
+    if not input_xyz.exists():
+        msg = f"Input file not found: {input_xyz}"
+        raise FileNotFoundError(msg)
+
+    positions = _collect_positions(input_xyz=input_xyz)
+
+    if not {"frame", "x", "y", "z"}.issubset(positions.columns):
+        msg = "Error in the .xyz format. Each line must be <x> <y> <z>."
+        raise ValueError(msg)
+
+    linked = tp.link_df(
+        positions,
+        search_range=search_range,
+        memory=memory,
+        adaptive_step=adaptive_step,
+        adaptive_stop=adaptive_stop,
+    )
+
+    with output_xyz.open("w") as f:
+        for frame_num in sorted(linked["frame"].unique()):
+            frame_data = linked[linked["frame"] == frame_num]
+            f.write(f"{len(frame_data)}\n")
+            f.write(f"Frame {frame_num}\n")
+            for _, row in frame_data.iterrows():
+                pid = int(row["particle"])
+                x, y, z = row["x"], row["y"], row["z"]
+                f.write(f"{pid} {x:.6f} {y:.6f} {z:.6f}\n")
+
+    logger.info(f"Linked .xyz file written to: {output_xyz}")
+    return Trj.init_from_xyz(traj_file=output_xyz, dt=1)
+
+
+def _collect_positions(input_xyz: Path) -> pd.DataFrame:
+    """Read the xyz file and return the positions dataset at each frame."""
+    lines = input_xyz.read_text().splitlines()
+
+    data = []
+    frame = -1
+    row = 0
+    dimensions = 3
+    for _ in range(len(lines)):
+        if row >= len(lines):
+            break
+        if lines[row].strip().isdigit():
+            num_atoms = int(lines[row])
+            frame += 1
+            row += 2  # skip comment line.
+            for a in range(num_atoms):
+                if row + a >= len(lines):
+                    break
+                parts = lines[row + a].strip().split()
+                if len(parts) >= dimensions:
+                    x, y, z = map(float, parts[0:3])
+                    data.append({"frame": frame, "x": x, "y": y, "z": z})
+            row += num_atoms
+        else:
+            row += 1
+
+    return pd.DataFrame(data)
diff --git a/src/dynsight/_internal/trajectory/trajectory.py b/src/dynsight/_internal/trajectory/trajectory.py
index 2f366d7d..23e77cdd 100644
--- a/src/dynsight/_internal/trajectory/trajectory.py
+++ b/src/dynsight/_internal/trajectory/trajectory.py
@@ -23,7 +23,7 @@
 class Insight:
     """Contains an analysis perfomed on a trajectory.
 
-    Attributes:
+    Parameters:
         dataset: The values of a some trajectory's descriptor.
         meta: A dictionary containing the relevant parameters.
     """
@@ -246,7 +246,7 @@ def get_onion_analysis(
 class ClusterInsight:
     """Contains a clustering analysis.
 
-    Attributes:
+    Parameters:
         labels: The labels assigned by the clustering algorithm.
     """
 
@@ -278,7 +278,7 @@ def load_from_json(cls, file_path: Path) -> ClusterInsight:
 class OnionInsight(ClusterInsight):
     """Contains an onion-clustering analysis.
 
-    Attributes:
+    Parameters:
         labels: The labels assigned by the clustering algorithm.
         state_list: List of the onion-clustering Gaussian states.
         reshaped_data: The input data reshaped for onion-clustering.
@@ -535,7 +535,7 @@ def plot_sankey(
 class Trj:
     """Contains a trajectory.
 
-    Attributes:
+    Parameters:
         universe: a MDAnalysis.Universe containing the trajectory.
 
     .. warning::
@@ -621,6 +621,7 @@ def get_soap(
         n_max: int,
         l_max: int,
         respect_pbc: bool = True,
+        selection: str = "all",
         centers: str = "all",
         n_core: int = 1,
     ) -> Insight:
@@ -635,6 +636,7 @@ def get_soap(
             soapnmax=n_max,
             soaplmax=l_max,
             soap_respectpbc=respect_pbc,
+            selection=selection,
             centers=centers,
             n_core=n_core,
         )
@@ -643,6 +645,7 @@ def get_soap(
             "n_max": n_max,
             "l_max": l_max,
             "respect_pbc": respect_pbc,
+            "selection": selection,
             "centers": centers,
         }
         return Insight(dataset=soap, meta=attr_dict)
diff --git a/src/dynsight/_internal/vision/vision_gui.py b/src/dynsight/_internal/vision/vision_gui.py
index d15b4ec0..bed2ca73 100644
--- a/src/dynsight/_internal/vision/vision_gui.py
+++ b/src/dynsight/_internal/vision/vision_gui.py
@@ -6,9 +6,11 @@
 from pathlib import Path
 from typing import Any
 
-from PIL import Image
+from PIL import Image, ImageTk
+from PIL.Image import Resampling
 
 
+# Dataclass for storing bounding box information.
 @dataclass
 class Box:
     id: int
@@ -19,12 +21,8 @@ class Box:
     abs_coords: tuple[int, int, int, int]
 
 
+# GUI class for creating synthetic datasets from images.
 class VisionGUI:
-    """GUI for interactively labeling images by drawing bounding boxes.
-
-    * Author: Simone Martino
-    """
-
     def __init__(
         self,
         master: tk.Tk,
@@ -35,47 +33,30 @@ def __init__(
         self.image_path = image_path
         self.destination_folder = destination_folder
         self.master.title("Dynsight: Label tool")
-        try:
-            self.image = tk.PhotoImage(file=image_path)
-        except Exception as e:
-            msg = f"Error loading image: {e}"
-            raise ValueError(msg) from e
-        # Setup the main grid
+
+        # Setup the GUI window size.
+        screen_width = self.master.winfo_screenwidth()
+        screen_height = self.master.winfo_screenheight()
+        self.master.geometry(f"{screen_width // 2}x{screen_height // 2}")
+
+        # Load the image.
+        pil_image = Image.open(self.image_path)
+        self.original_image = pil_image
+        self.original_width, self.original_height = pil_image.size
+
+        self.scale = 1.0
+        self.canvas_image_id: int | None = None
+
+        # GUI window layout.
         self.master.rowconfigure(index=0, weight=1)
-        # Image
         self.master.columnconfigure(index=0, weight=1)
-        # Sidebar
-        self.master.columnconfigure(index=1, weight=1)
+        self.master.columnconfigure(index=1, weight=0)
 
-        # Image canvas
-        self.canvas = tk.Canvas(
-            master=self.master,
-            width=self.image.width(),
-            height=self.image.height(),
-            cursor="crosshair",
-        )
+        # Canvas for image labelling.
+        self.canvas = tk.Canvas(master=self.master, cursor="crosshair")
         self.canvas.grid(row=0, column=0, sticky="nsew")
-        self.canvas.create_image(0, 0, anchor=tk.NW, image=self.image)
-        # Rulers
-        self.h_line = self.canvas.create_line(
-            0,  # x0
-            0,  # y0
-            self.image.width(),  # x1
-            0,  # y1
-            fill="blue",
-            dash=(2, 2),
-            width=3,
-        )
-        self.v_line = self.canvas.create_line(
-            0,  # x0
-            0,  # y0
-            0,  # x1
-            self.image.height(),  # y1
-            fill="blue",
-            dash=(2, 2),
-            width=3,
-        )
-        # Sidebar
+
+        # Buttons.
         self.sidebar = tk.Frame(
             master=self.master,
             width=150,
@@ -85,7 +66,6 @@ def __init__(
         self.sidebar.grid(row=0, column=1, sticky="ns")
         self.sidebar.grid_propagate(flag=False)
 
-        # Buttons
         self.submit_button = tk.Button(
             self.sidebar,
             text="Submit",
@@ -107,21 +87,105 @@ def __init__(
         )
         self.close_button.pack(pady=10, fill="x")
 
-        # Labelling variables
         self.start_x = 0
         self.start_y = 0
         self.current_box = 0
         self.boxes: list[Box] = []
 
-        # Mouse bindings
+        # Rulers.
+        self.h_line = self.canvas.create_line(
+            0,  # x0
+            0,  # y0
+            0,  # x1
+            0,  # y1
+            fill="blue",
+            dash=(2, 2),
+            width=3,
+        )
+        self.v_line = self.canvas.create_line(
+            0,  # x0
+            0,  # y0
+            0,  # x1
+            0,  # y1
+            fill="blue",
+            dash=(2, 2),
+            width=3,
+        )
+
+        # Bindings.
         self.canvas.bind("<Button-1>", self._on_click_press)
         self.canvas.bind("<ButtonRelease-1>", self._on_click_release)
         self.canvas.bind("<B1-Motion>", self._on_mouse_drag)
         self.canvas.bind("<Motion>", self._follow_mouse)
+        self.master.bind("<Configure>", self._resize)
+
+        # Initial rendering of the image.
+        self._resize()
+
+    # Resize image and redraw everything when the window is resized
+    def _resize(self, _: tk.Event[Any] | None = None) -> None:
+        canvas_width = self.canvas.winfo_width()
+        canvas_height = self.canvas.winfo_height()
+        size_thr = 10
+        if canvas_width < size_thr or canvas_height < size_thr:
+            return
+
+        # Scaling factor.
+        self.scale = min(
+            canvas_width / self.original_width,
+            canvas_height / self.original_height,
+        )
+        resized = self.original_image.resize(
+            (
+                int(self.original_width * self.scale),
+                int(self.original_height * self.scale),
+            ),
+            Resampling.LANCZOS,
+        )
+        # Update image and boxes.
+        self.tk_image = ImageTk.PhotoImage(resized)
+        self.canvas.delete("all")
+        self.canvas_image_id = self.canvas.create_image(
+            0,
+            0,
+            anchor=tk.NW,
+            image=self.tk_image,
+        )
+
+        # Re-draw boxes
+        for box in self.boxes:
+            x1, y1, x2, y2 = [int(v * self.scale) for v in box.abs_coords]
+            box.id = self.canvas.create_rectangle(
+                x1,
+                y1,
+                x2,
+                y2,
+                outline="red",
+                width=3,
+            )
 
-    # Mouse functions
+        # Recreate guide lines
+        self.h_line = self.canvas.create_line(
+            0,  # x0
+            0,  # y0
+            canvas_width,  # x1
+            0,  # y1
+            fill="blue",
+            dash=(2, 2),
+            width=3,
+        )
+        self.v_line = self.canvas.create_line(
+            0,  # x0
+            0,  # y0
+            0,  # x1
+            canvas_height,  # y1
+            fill="blue",
+            dash=(2, 2),
+            width=3,
+        )
+
+    # Start drawing a box.
     def _on_click_press(self, event: tk.Event[Any]) -> None:
-        """Starts drawing the box on mouse press."""
         self.start_x, self.start_y = event.x, event.y
         self.current_box = self.canvas.create_rectangle(
             self.start_x,  # x0
@@ -132,15 +196,23 @@ def _on_click_press(self, event: tk.Event[Any]) -> None:
             width=3,
         )
 
+    # Finish drawing the box and store its data
     def _on_click_release(self, event: tk.Event[Any]) -> None:
-        """Finalize the box on mouse release."""
         x2, y2 = event.x, event.y
         x1, y1 = self.start_x, self.start_y
-        abs_coords = (min(x1, x2), min(y1, y2), max(x1, x2), max(y1, y2))
-        center_x = (x1 + x2) / (2 * self.image.width())
-        center_y = (y1 + y2) / (2 * self.image.height())
-        width_rel = abs(x2 - x1) / self.image.width()
-        height_rel = abs(y2 - y1) / self.image.height()
+        x1_orig = int(min(x1, x2) / self.scale)
+        y1_orig = int(min(y1, y2) / self.scale)
+        x2_orig = int(max(x1, x2) / self.scale)
+        y2_orig = int(max(y1, y2) / self.scale)
+
+        abs_coords = (x1_orig, y1_orig, x2_orig, y2_orig)
+
+        center_x = (x1_orig + x2_orig) / (2 * self.original_width)
+        center_y = (y1_orig + y2_orig) / (2 * self.original_height)
+        width_rel = abs(x2_orig - x1_orig) / self.original_width
+        height_rel = abs(y2_orig - y1_orig) / self.original_height
+
+        # Create and store Box object.
         box_info = Box(
             id=self.current_box,
             center_x=center_x,
@@ -151,54 +223,40 @@ def _on_click_release(self, event: tk.Event[Any]) -> None:
         )
         self.boxes.append(box_info)
         self.current_box = 0
+        self._resize()  # Redraw for consistent scaling
 
     def _on_mouse_drag(self, event: tk.Event[Any]) -> None:
-        """Update the box coordinates while dragging the mouse."""
         sel_x, sel_y = event.x, event.y
         self.canvas.coords(
-            self.current_box,  # ID
+            self.current_box,
             self.start_x,  # x0
             self.start_y,  # y0
             sel_x,  # x1
             sel_y,  # y1
         )
-        # Sync rulers too
         self.canvas.coords(
-            self.h_line,  # ID
+            self.h_line,
             0,  # x0
             sel_y,  # y0
-            self.image.width(),  # x1
+            self.canvas.winfo_width(),  # x1
             sel_y,  # y1
         )
         self.canvas.coords(
-            self.v_line,  # ID
+            self.v_line,
             sel_x,  # x0
             0,  # y0
             sel_x,  # x1
-            self.image.height(),  # y1
+            self.canvas.winfo_height(),  # y1
         )
 
+    # Update box as mouse is dragged.
     def _follow_mouse(self, event: tk.Event[Any]) -> None:
-        """Sync guide lines position with mouse movement."""
         x, y = event.x, event.y
-        self.canvas.coords(
-            self.h_line,  # ID
-            0,  # x0
-            y,  # y0
-            self.image.width(),  # x1
-            y,  # y1
-        )
-        self.canvas.coords(
-            self.v_line,  # ID
-            x,  # x0
-            0,  # y0
-            x,  # x1
-            self.image.height(),  # y1
-        )
+        self.canvas.coords(self.h_line, 0, y, self.canvas.winfo_width(), y)
+        self.canvas.coords(self.v_line, x, 0, x, self.canvas.winfo_height())
 
-    # Button functions
+    # Save cropped images from each bounding box.
     def _submit(self) -> None:
-        """Save the cropped images."""
         cropped_img_folder = self.destination_folder / "training_items"
         cropped_img_folder.mkdir(exist_ok=True)
         pil_image = Image.open(self.image_path)
@@ -209,12 +267,13 @@ def _submit(self) -> None:
             cropped_image.save(save_path)
         self.master.destroy()
 
+    # Remove last drawn box.
     def _undo(self) -> None:
-        """Remove the last drawn box."""
         if self.boxes:
             last_box = self.boxes.pop()
             self.canvas.delete(last_box.id)
+            self._resize()
 
+    # Close the GUI
     def _close(self) -> None:
-        """Close without saving."""
         self.master.destroy()
diff --git a/src/dynsight/track.py b/src/dynsight/track.py
new file mode 100644
index 00000000..b88355f5
--- /dev/null
+++ b/src/dynsight/track.py
@@ -0,0 +1,9 @@
+"""track package."""
+
+from dynsight._internal.track.track import (
+    track_xyz,
+)
+
+__all__ = [
+    "track_xyz",
+]