A Geometric View Transformation Model Using Free-Form Deformation for Cross-View Gait Recognition

Author(s):  
Hazem El-Alfy ◽  
Chi Xu ◽  
Yasushi Makihara ◽  
Daigo Muramatsu ◽  
Yasushi Yagi
2010 ◽  
Author(s):  
Nicholas J. Tustison ◽  
Suyash Awate ◽  
James Gee

Our previous contributions to the ITK community include a generalized B-spline approximation scheme as well as a generalized information-theoretic measure for assessing point-set correspondence known as the Jensen-Havrda-Charvat-Tsallis (JHCT) divergence. In this submission, we combine these two contributions for the registration of labeled point-sets. The transformation model which uses the former contribution is denoted as directly manipulated free-form deformation (DMFFD) and has been used for image registration. The information-theoretic approach described not only eliminates exact cardinality constraints which plague exact landmark matching algorithms, but it also incorporates the local point-set structure into the similarity measure calculation. Although theoretical discussion of these two components is deferred to other venues, the implementation details given in this submission should be adequate for those wishing to use our algorithm. Visualization of results is aided by another of our previous contributions. Additionally, we provide the rudimentary command line parsing classes used in our testing routines which were written in the ITK style and also available to use consistent with the open-source paradigm.


Author(s):  
Yasushi Makihara ◽  
Ryusuke Sagawa ◽  
Yasuhiro Mukaigawa ◽  
Tomio Echigo ◽  
Yasushi Yagi

Author(s):  
MAODI HU ◽  
YUNHONG WANG ◽  
ZHAOXIANG ZHANG

Considering it is difficult to guarantee that at least one continuous complete gait cycle is captured in real applications, we address the multi-view gait recognition problem with short probe sequences. With unified multi-view population hidden markov models (umvpHMMs), the gait pattern is represented as fixed-length multi-view stances. By incorporating the multi-stance dynamics, the well-known view transformation model (VTM) is extended into a multi-linear projection model in a four-order tensor space, so that a view-independent stance-independent identity vector (VSIV) can be extracted. The main advantage is that the proposed VSIV is stable for each subject regardless of the camera location or the sequence length. Experiments show that our algorithm achieves encouraging performance for cross-view gait recognition even with short probe sequences.


Author(s):  
Stavros N. Leloudas ◽  
Giorgos A. Strofylas ◽  
Ioannis K. Nikolos

Given the importance of structural integrity of aerodynamic shapes, the necessity of including a cross-sectional area equality constraint among other geometrical and aerodynamic ones arises during the optimization process of an airfoil. In this work an airfoil optimization scheme is presented, based on Area-Preserving Free-Form Deformation (AP FFD), which serves as an alternative technique for the fulfillment of a cross-sectional area equality constraint. The AP FFD is based on the idea of solving an area correction problem, where a minimum possible offset is applied on all free-to-move control points of the FFD lattice, subject to the area preservation constraint. Due to the linearity of the area constraint in each axis, the extraction of an inexpensive closed-form solution to the area preservation problem is possible by using Lagrange Multipliers. A parallel Differential Evolution (DE) algorithm serves as the optimizer, assisted by two Artificial Neural Networks as surrogates. The use of multiple surrogate models, in conjunction with the inexpensive solution to the area correction problem, render the optimization process time efficient. The application of the proposed methodology for wind turbine airfoil optimization demonstrates its applicability and effectiveness.


2013 ◽  
Vol 60 (3) ◽  
pp. 537-563 ◽  
Author(s):  
Francesco Ballarin ◽  
Andrea Manzoni ◽  
Gianluigi Rozza ◽  
Sandro Salsa

Author(s):  
Vincent Cheutet ◽  
Jean-Philippe Pernot ◽  
Jean-Claude Leon ◽  
Bianca Falcidieno ◽  
Franca Giannini

To limit low-level manipulations of free-form surfaces, the concept of Fully Free Form Deformation Features (δ-F4) have been introduced. They correspond to shapes obtained by deformation of a surface area according to specified geometric constraints. In our work, we mainly focused on those features aimed at enforcing the visual effect of the so-called character lines, extensively used by designers to specify the shape of an object. Therefore, in the proposed approach, 3D lines are used to drive surface deformation over specified areas. Depending on the wished shape and reflection light effects, the insertion of character lines may generate surface tangency discontinuities. In CAD systems, such kind of discontinuities is generally created by a decomposition of the initial surface into several patches. This process can be tedious and very complex, depending on the shape of the deformation area and the desired surface continuity. Here, a method is proposed to create discontinuities on a surface, using the trimming properties of surfaces. The corresponding deformation features produce the resulting surface in a single modification step and handle simultaneously more constraints than current CAD systems. The principle of the proposed approach is based on arbitrary shaped discontinuities in the parameter domain of the surface to allow the surface exhibiting geometric discontinuities at user-prescribed points or along lines. The proposed approach is illustrated with examples obtained using our prototype software.


Sign in / Sign up

Export Citation Format

Share Document