“Let us walk around 3D virtual cities as if we were there!”

Morphology applied to urban modeling


Urban modeling is the keystone for the development of many applications such as: 3D navigation, urban planning, scene modeling for the entertainment industry, etc. The technological challenge is to increase the productivity and the realism of urban modeling. The PhD thesis of Jorge Hernandez has been completed at the Centre for Mathematical Morphology of MINES ParisTech, in partnership with Cap Digital Business Cluster TerraNumerica project (consortium of 19 private and public organisations).

 Partnership for a project

The PhD thesis of Jorge Hernandez has been completed at the Centre for Mathematical Morphology of MINES ParisTech, in partnership with Cap Digital Business Cluster TerraNumerica project (consortium of 19 private and public organisations).

 

This particular project aims at developing a production and exploitation platform, by enabling the definition and visualisation of synthetic urban scenes.  Two areas have been chosen:

- The first one concerns the façade analysis of Haussmann's Architectural Paris. Façade modeling is performed on a single building. However, during the image acquisition process, several buildings appear in the same image. We have developed an automatic method to separate different façades included in the image. We then focused on semantic extraction from the façade. It consists in segmenting it by floors, windows, balconies... to provide a realistic building model.

- The second one concerned point cloud analysis. In urban modeling, it is necessary to introduce some elements such as street furniture and pavement. We present tools for the detection and classification of artifacts. These tools allow: 1- the filtering of data in order to facilitate the modeling process and 2- the re-introduction of some elements (lampposts, sign boards, bus stop, etc), improving visual realism in the modeled scenes. We also propose an automatic method for pavement segmentation.

For more details http://www.competitivite.gouv.fr/spip.php?article636

                                                               

Jorge wanted to combine in his PhD thesis his favorite research areas, image processing methods applied to 3D modeling. He chose Mines ParisTech, because he considered it is one of France top-notch engineering colleges and frontrunner in the field of science and technology.

Jorge Hernandez was born in Manizales (one of the most important Coffee Regions of Colombia), in 1982. He received a B.Sc. degree in Electronic Engineering from National University of Colombia in 2004. In his degree thesis, he worked on classification of coffee beans during the drying process using image processing algorithms. 

This experience instilled in him the desire to pursue his studies in the image processing field.  In 2006, he received a M.Sc. degree in Engineering from the same institution. His master thesis consisted in point cloud reconstruction via stereo image and structured light methods (computer vision) and texture mapping and 2D/3D registration in order to model Pre-Columbian artefacts (Cultural Heritage Applications). In addition, he has had an internship experience at Ampère Laboratory, INSA LYON within the framework of ECOS-NORD France Colombia Cooperation project working on human machine interface and video tracking.

Deeply interested in the field of Image Processing and Computer Vision, Jorge submitted an application for Doctoral studies at Mines-ParisTech. In his PhD, he developed innovative filtering and segmentation algorithms based on Mathematical Morphology.


Jorge is currently working as Image Science Engineer in the DxOLabs developing novel denoising algorithms.