Large urban agglomerations nowadays are facing some major issues such as economic restrictions, environmental challenges, global and systemic approaches in city management [8]. One of them is to precisely monitor urban objects which can be natural (trees), artificial (traffic lights or poles), static or moving (cars). This is essential to analyze their mutual interaction (for example branches of a tree which are close to an electric pole) and to prevent risks associated with them (for example, dead parts of a tree which may fall on a street). Thus being able to localize urban objects and provide informations about their status is essential. One way of achieving this task is to mount LiDAR scanners on vehicles and analyze acquisitions performed on a regular time basis. However this requires to automatically process the 3D point clouds given by the LiDAR. Automatic object detection has seen an increase of popularity in the recent years due to the rise of Deep-Learning methods [3] being able to achieve human-like performances for object classification in 2D images. This increase of performance is starting to have a huge impact on research in 3D shape recognition (see papers presented on the Web site). Given a complete scene obtained with a LiDAR scanner, i.e. a cloud of 3D points, the ultimate goal is then to localize and identify 3D urban objects. In this paper, we focus on the classification of small part of the 3D scene reduced to a single object. This requires to use a new Deep Learning method dedicated to 3D points. In the following, we briefly review existing methods and we present some preliminary results obtained with the PointNet network.