UrTrees is a citizen science project aimed at studying urban trees. For decades, foresters have collected data on trees in their natural environment and used this data to build growth and gas exchange models. These models are now used to predict the influence of forest areas on climate change and find new solutions to this threat. However, for now, we know little about urban trees that grow differently due to different soil compositions, different air quality, regular pruning, etc. Collecting a large amount of data on urban trees is therefore a crucial step to include these trees and their ecosystem benefits (natural habitat for urban animal species, reduction of heat zones, improvement of quality of life, etc.). However, time is running out, and we decided to call on citizens' participation to fill the data gap on such trees.

  Collecting a large amount of measurements on urban trees in different locations is a long and tedious process. UrTrees was designed to allow any citizen to help researchers collect this data in a simple and fast way. For example, in a single day, a dedicated operator can collect data on 20 trees, while the same amount of data can be collected in twenty minutes by a group of ten citizens without prior training. In addition, UrTrees allows citizens around the world to collect data around them, offering a wide range of measurements in various locations.

  By using this application, you play an important role in our research, and we would like to thank you. When you submit a tree video through UrTrees, we process it (this can take about ten minutes) and then provide you with:
   • Of course, the measurements of the tree you are sending us a video of.
   • The most likely species (if you have also provided a photo of a leaf or the bark of the trunk).
   • The tree virtually reconstructed in 3D as a point cloud.
   • Some points, allowing you to complete quests and climb the contributors' ranking, which you can share and which allows you to challenge other users! On this website, you also have access to all the contributions from around the world.

  A few metres: not too close to be able to film up to the top of the tree, not too far to avoid missing details (depends on the resolution of your video). The exact distance is not critical and can vary as you walk around the tree.

  No, it is not necessary. It is preferable because it allows to have more details and therefore a more accurate estimation of the three computed features, but doing a little less than half a turn is generally enough.

  One ​​to two minutes is a reasonable duration. A shorter video may not contain enough information to accurately estimate the tree features. A longer video will take longer to process.

  These weather conditions should be avoided: wind will move the tree branches, which will affect the accuracy of the feature estimates. Similarly, rain can have a negative impact on the sharpness of your video and therefore the accuracy of the estimation.

  The automatic processing of each video takes about ten minutes. However, if other users send their videos to the server at the same time as you, these videos are processed one after the other. It is therefore possible that you will receive the results of your video several dozen minutes after sending it to our server. If you did not get these results and think there was a bug, try resubmitting your video.

  Here is an overview of how your data is processed. The whole process is fully automatic.
   • Images are extracted from your video.
   • These images are processed by a so-called Structure-from-Motion algorithm that builds a 3D point cloud from them.
   • This point cloud is processed by different algorithms of our invention to isolate the points corresponding to the target tree from the ground, neighbouring trees, street furniture, etc.
   • 3D geometric models are fitted to the points corresponding to the trunk and crown of the tree respectively, in order to derive the three measurements that interest us.

  UrTrees currently measures three main features of trees:
   • The diameter at breast height of the tree (DBH), which is the diameter of the trunk at 1.30 metres above the ground.
   • The height of the tree, from the ground to its crown.
   • The volume of the crown of the tree. More features may be added in the future. Feel free to send us your suggestions!

  No! We do not store any of your personal information, do not sell it and never will!
First of all, the only personal data we collect (pseudonym, height and location at the time of video capture) are completely anonymous and do not normally allow you to be identified. This data is used only for two scientific purposes: 1 Your height is only required to allow accurate scaling of the 3D point clouds reconstructed from your videos. It is stored permanently only on your smartphone, not on a remote server. 2 The location of the tree you scan is only required to precisely identify it and to be able to track it over time. As for the pseudonym, it just lets us know what your contributions are and your ranking :)