Human Body Reconstruction from Image Sequences

Human Body Reconstruction from Image Sequences,10.1007/3-540-45783-6_7,Fabio Remondino

Human Body Reconstruction from Image Sequences   (Citations: 2)
BibTex | RIS | RefWorks Download
The generation of 3-D models from uncalibrated sequences is a challenging problem that has been investigated in many research activities in the last decade. In particular, a topic of great interest is the modeling of real humans. In this paper a method for the 3-D reconstruction of static human body shapes from images acquired with a video-camera is presented. The process includes the orientation and calibration of the sequence, the extraction of correspondences on the body using least squares matching technique and the reconstruction of the 3-D point cloud of the human body. The actual interests in 3-D object reconstruction are motivated by a wide spectrum of applications, such as object recognition, city modeling, video games, animations, surveillance and visualization. In the last years, great progress in creating and visualizing 3-D models of man-made and non-rigid objects from images has been made, with particular attention to the visual quality of the results. The existing systems are often built around specialized hardware (laser scanner), resulting in high costs. Other methods based on photogrammetry (12, 19) or computer vision (18), can instead obtain 3-D models of objects with low cost acquisition systems, using photo or video cameras. Since many years, photogrammetry deals with high accuracy measurements from image sequences, including 3-D object tracking (15), deformation measurements or motion analysis (7); these applications requires very precise calibration, but automated and reliable procedures are available. In the last years many researchers have tried to increase the automation level for the production of 3-D models reducing the requirements and the accuracy of the calibration, with the goal of automatically extract a 3-D model "by freely moving a camera around an object" (18). Concerning the reconstruction and modeling of human bodies, nowadays the demand for 3-D models has drastically increased. A complete model of human consists of both the shape and the movements of the body. Many available systems consider these two modeling processes as separate even if they are very close. A classical approach to build human shape models uses 3-D scanners (Cyberware): they are expensive but simple to use and various modeling software are available. Other techniques use structured light methods (2), silhouette extraction (23) and multi- image photogrammetry (6). These human models can be used for different purposes, like animation, manufacture or medical applications. For animation, only approximative measurements are necessary: the shape can be first defined (e.g. with
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
Sort by: