Visual Registration with Physically Coherent Models
Visual Registration with Physically Coherent Models

Physically coherent modeling, accurate registration and development of associated robust estimation techniques are  the main focus of the tangram team.

Visual registration and modeling are two  research topics with a rich history in computer vision. The TANGRAM project aims at addressing some specific aspects of these fields which are still largely unsolved. Indeed,  difficulties originate in the nature of the scene (poorly textured or specular environments), in the  motion undergone by the object (tiny motions which hardly emerge from the noise floor, or in contrast, highly deformable objects) and in dissimilarities which may occur in the scene between the time the modeling stage occurs and the application time. In the continuity of the MAGRIT team, applications will be mostly dedicated to interactive tasks but are not limited to these aspects.

Registration and modeling are two joined topics  and a tradeoff between physical  realism of the model, convergence issues and robusness of the registration or tracking  tasks must be achieved. Recently, the ability of neural networks to learn feature descriptors,  to aggregate information of various complexities in images and to reason at the object level  has opened new and promising  perspectives both for  modeling and for registration.  While continuing to explore these problems with more classical approaches derived from  signal processing, geometry and robust esimation, one of the aim of the team is to integrate machine learning methods  into 3D geometric tasks.

The TANGRAM project is structured along three lines which aim at:

  • better integrating machine learning techniques into the geometric problems under consideration.
  • building physically coherent models with a good accuracy versus efficiency compromise fort the targeted tasks. The models considered here range from classical geometric models to the design of new models built at the object level or augmented with semantic information.
  • designing robust numerical schemes for solving ill posed problems raised by the above questions.

Targeted trans-disciplinary applications  are  mixed and augmented  reality, computational photomechanics and minimally invasive medical interventions.

Centre(s) inria
Inria centre at Université de Lorraine
In partnership with
Université de Lorraine,CNRS


Team leader

Véronique Constant

Team assistant

Sylvie Hilbert

Team assistant

Antoinette Courrier

Team assistant

Juline Brevillet

Team assistant

Nathalie Bethus

Team assistant