MIIRE — Multimedia, Image, Indexing and Recognition

Research activities

The main focus of the MIIRE research group is on statistical 3D shape analysis and its applications in a wide range of fields such as: (1) segmentation, matching, and retrieval of 3D data, (2) face and facial expression recognition from data acquired with 3D scanning devices, and (3) 3D-data acquisition.

Segmentation, matching and retrieval of 3D data

The use of three-dimensional image and model databases throughout the Internet is growing both in number and in size. The development of modeling tools, 3D-scanners, 3D-graphic accelerated hardware, Web3D, and so on, is enabling access to 3D materials of high quality. In recent years, many systems have been proposed for efficient information retrieval, segmenting and analysis from digital collections of images and videos. However, the solutions proposed so far are not always effective in application contexts where the information is intrinsically 3D.

The goal of this research topic is to develop a new geometrical and statistical tools for segmenting, matching, deforming, comparing, and averaging shapes of surfaces.

Graph extraction and segmentation of a 3D-model – CAD-models – 3d-search web engine.

3D-face and facial expression recognition

There is an increasing interest in analyzing shapes of facial surfaces with many applications including biometrics, facial surgery, video communications, and 3-D animation. This interest is fuelled by the advent of cheaper and lighter scanners that can provide high-resolution measurements of both geometry and texture of human facial surfaces.

One general goal here is to develop computational tools for analyzing 3D face data.

Examples of intra and inter-class geodesic path between faces.

3D-data acquisition

Nowadays, significant advances have been made in the field of 3D television technologies. Many manufacturers are providing hardware solutions for immersive 3D visualizations, and 3DTV sets are starting to equip our living rooms. While the 3D visualization technologies are evolving at a fast paste, technologies for 3D content management, such as acquisition, storage, streaming, and watermarking, are still far behind their 2D counter-parts. Despite the substantial amount of papers that have been published in this field over the past years, several open challenges remain unsolved, particularly when it comes to capturing and reproducing highly detailed 3D video of dynamic scenes with complex material properties and at video rates.

In our research group we investigate and develop new methodologies for bringing the 3DTV technologies to the end-user level. Particularly, we focus on the 3D acquisition and contents creation process, and investigate new hardware as well as new algorithmic solutions.