Facade Segmentation


We focus on a segmentation procedure behind the facade modeling to detect/extract structural objects, mainly windows. This facade analysis is developed in the framework of Cap Digital Business Cluster TerraNumerica project. This project aims at developing a production and exploitation platform, by allowing the definition and visualization of synthetic urban environments. The platform aims at increasing the productivity and the realism of urban modeling.

Scene-Text Detection

The text in a scene is linked to the semantic context of the image and constitutes a relevant descriptor for content-based image indexing. This research is developed as part of iTOWNS (Image-based Town On-line Web navigation and Search Engine) project. The goal of this project is to develop a new generation of multimedia web tools that combine a 3D geographic browser with an image-based search engine based on a data indexing images and visual context.