Section: Bilateral Contracts and Grants with Industry

Bilateral Contracts with Industry

Google

Participants : Pierre Alliez, Cedric Portaneri.

We aim to develop a novel approach and software prototype for the compression of 3D models. Our main focus is on progressive compression of surface triangle meshes with color textures, with emphasis on fine grain, genericity and flexible metric. The proposed methodology is to turn the input models into a stream of refinements, in which both mesh and texture refinement details are multiplexed in accordance to rate-distortion principles. Fine grain will be achieved through considering all components, local as well as non-local, from both the mesh and its textures: mesh complexity, vertex accuracy, texture definition and accuracy. We will leverage the recent advances on perceptual metrics to improve the visual appearance, and perform joint consolidation and encoding of the models to further optimize the rate-distortion tradeoffs and visual perception.

- Starting date: January 2017 - Duration: 1 year

Geoimage

Participants : Liuyun Duan, Florent Lafarge.

The aim of this collaboration is to devise a new type of 2.5D representation from satellite multi-view stereo images which is more accurate, compact and meaningful than the conventional digital elevation models (DEMs). A key direction consists in incorporating semantic information directly during the image matching process. Such a semantic information is related to the type of components of the scene, such as vegetation, roofs, building edges, roads and land.

- Starting date: November 2013 - Duration: 4 years

CSTB

Participants : Hao Fang, Florent Lafarge.

The goal of this recent collaboration is to develop methods for analyzing and exploring scale-spaces into urban 3D data.

- Starting date: March 2016 - Duration: 3 years

Luxcarta

Participants : Jean-Philippe Bauchet, Florent Lafarge.

The goal of this recent collaboration is to design automated approaches for producing city models from the last generation of satellites. The models should conform to the level 2 (LOD2) of the popular CityGML format.

- Starting date: October 2016 - Duration: 3 years

CNES

Participants : Emmanuel Maggiori, Yuliya Tarabalka.

The objective of the project was to devise hierarchical approaches for object-oriented classification of multi-source images. Multi-source images are generated from a scene observed by different types of sensors.

- Starting date: November 2015 - Duration: 2 years

CNES and Acri-ST

Participants : Onur Tasar, Yuliya Tarabalka.

The aim is to devise efficient representations for satellite images.

- Starting date: October 2017 - Duration: 3 years