diff --git a/Multi-Session-Mapping-with-RTAB-Map-Tango.md b/Multi-Session-Mapping-with-RTAB-Map-Tango.md
index c80be6d..547dea9 100644
--- a/Multi-Session-Mapping-with-RTAB-Map-Tango.md
+++ b/Multi-Session-Mapping-with-RTAB-Map-Tango.md
@@ -1,6 +1,8 @@
 ## Introduction
 This tutorial shows how to merge multiple RTAB-Map databases taken with [RTAB-Map Tango](https://play.google.com/store/apps/details?id=com.introlab.rtabmap). The same approach would apply with databases taken with RTAB-Map Desktop and `rtabmap_ros`. To see a multi-session example from data captured on a robot, see the original [Multi-Session tutorial](https://github.com/introlab/rtabmap/wiki/Multi-session).
 
+[[doc/Tutorials/MultiSessionTango/cloud.jpg]][[doc/Tutorials/MultiSessionTango/cloudSessionColor.jpg]]
+
 ## Steps
 
 ### 1. Scanning tips
@@ -53,6 +55,7 @@ To optimize the map even more, click on Tools->Post-processing, check Detect mor
 ### 7. Export the mesh. 
 Click "File->Export 3D clouds..." and setup the panel as the following screenshot and click "Save". If you prefer exporting the point cloud and create the mesh (and texturing) with another software, you can see the [Export Rasters Layer to MeshLab](https://github.com/introlab/rtabmap/wiki/Export-Raster-Layers-to-MeshLab) tutorial. Wait a couple of minutes then you will be able to save the OBJ.
     [[doc/Tutorials/MultiSessionTango/export.jpg]]
+    [[doc/Tutorials/MultiSessionTango/mesh.jpg]]
 
 ## Other examples with Tango
 The following examples are single session mapping, but have very similar workflow than shown in this tutorial to re-process the database. In these cases, the databases have been re-processed with different parameters to get more loop closures than with the default parameters.