Filters








290 Hits in 1.8 sec

Expressive Gait Synthesis Using PCA and Gaussian Modeling [chapter]

Joëlle Tilmanne, Thierry Dutoit
2010 Lecture Notes in Computer Science  
In that reduced PC space, the variability of walk cycles for each emotion and the length of each cycle are modeled using Gaussian distributions.  ...  Using this modeling, new sequences of walk can be synthesized for each expression, taking into account the variability of walk cycles over time in a continuous sequence.  ...  A Gaussian modeling of the data represented in the PC space is then conducted and enables us to model the variability of the walk cycles over time and thus to introduce some randomness in the synthesized  ... 
doi:10.1007/978-3-642-16958-8_34 fatcat:yzrfgc3tfffzzp2ticwxz2guuu

Multi-layer joint gait-pose manifold for human motion modeling

Meng Ding, Guoliang Fan
2013 2013 10th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (FG)  
The experimental results confirm the advantages and superiority of our proposed method over several existing Gaussian process-based motion models.  ...  We present a multi-layer joint gait-pose manifold (multi-layer JGPM) for human motion modeling to enhance the representative capability of the original JGPM that represents gait kinematics by two variables  ...  constraint [6] and PCA are shown in Fig. 3 (a) and (b).  ... 
doi:10.1109/fg.2013.6553783 dblp:conf/fgr/DingF13 fatcat:2nmant5lkfh3hhjglq6padtjpy

Reactive Statistical Mapping: Towards the Sketching of Performative Control with Data [chapter]

Nicolas d'Alessandro, Joëlle Tilmanne, Maria Astrinaki, Thomas Hueber, Rasmus Dall, Thierry Ravet, Alexis Moinet, Huseyin Cakmak, Onur Babacan, Adela Barbulescu, Valentin Parfait, Victor Huguenin (+2 others)
2014 IFIP Advances in Information and Communication Technology  
The idea of statistical mapping has been investigated, more precisely how to use Gaussian Mixture Models and Hidden Markov Models for realtime and reactive generation of new trajectories from inputted  ...  Our target for this workshop was to bring some technologies currently used in speech recognition and synthesis to a new level, i.e. being the core of a new HMM-based mapping system.  ...  Gaussian Mixture Models (GMMs) and Hidden Markov Models (HMMs).  ... 
doi:10.1007/978-3-642-55143-7_2 fatcat:ewdc5fk33zbe5dgpon6jtaznju

Multifactor Gaussian process models for style-content separation

Jack M. Wang, David J. Fleet, Aaron Hertzmann
2007 Proceedings of the 24th international conference on Machine learning - ICML '07  
We demonstrate our approach using time-series prediction, and by synthesizing novel animation from the model.  ...  We learn models for human locomotion data, in which each pose is generated by factors representing the person's identity, gait, and the current state of motion.  ...  The data used in this project was obtained from mocap.cs.cmu.edu, which was created with funding from NSF EIA-0196217.  ... 
doi:10.1145/1273496.1273619 dblp:conf/icml/WangFH07 fatcat:f4h6yzrnwjgchkhaz7om7gnsbq

Fibre Centred Tensor Faces

B. P. Tiddeman, D. W. Hunter, Y. Meng
2007 Procedings of the British Machine Vision Conference 2007  
The new method is tested on image synthesis and analysis and in a simple face recognition task, in which it out performs non-centred multilinear analysis.  ...  In this paper we present a reformulation of the tensorface analysis method and produce a model that is simpler (i.e. has fewer parameters), is more compact (i.e. has tighter distributions) and is less  ...  We could populate the Gaussian model by using additional example images of the same subjects used to train the tensor model but as this requires an additional image set, we instead use the simpler method  ... 
doi:10.5244/c.21.29 dblp:conf/bmvc/TiddemanYH07 fatcat:goevwwzn4rahlhogc2bykzl64q

Classification and Recognition of Dynamical Models: The Role of Phase, Independent Components, Kernels and Optimal Transport

Alessandro Bissacco, Alessandro Chiuso, Stefano Soatto
2007 IEEE Transactions on Pattern Analysis and Machine Intelligence  
We validate our choice of models, inference algorithm, and distance on the tasks of human motion synthesis (sample paths of the learned models) and recognition (nearest-neighbor classification in the computed  ...  Once a model is identified, we define a kernel-based cord distance between models, which includes their dynamics, their initial conditions, and input distribution.  ...  The authors also wish to thank Rene Vidal for discussions and suggestions on the problem of comparing dynamical models, and specifically on the sue of kernels for this task.  ... 
doi:10.1109/tpami.2007.1101 pmid:17848777 fatcat:cnaxdr7pi5bubedivwtlxlom5u

Motion Models for People Tracking [chapter]

David J. Fleet
2011 Visual Analysis of Humans  
This chapter provides an introduction to models of human pose and motion for use in 3D human pose tracking.  ...  We concentrate on probabilistic latent variable models of kinematics, most of which are learned from motion capture data, and on recent physics-based models.  ...  [76] learned a three-factor model, using the subject's identity, the gait type (walk, stride or jog), and the phase of the gait cycle.  ... 
doi:10.1007/978-0-85729-997-0_10 fatcat:fr7d7dpvhfepvdwh4mkhfpcqby

Video from nearly still: An application to low frame-rate gait recognition

N. Akae, A. Mansur, Y. Makihara, Y. Yagi
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
A periodic image sequence is expressed as a manifold parameterized by a phase and a standard manifold is learned from multiple high frame-rate sequences in the training stage.  ...  The proposed method is applied to low frame-rate gait recognition and experiments with real data of 100 subjects demonstrate a significant improvement by the proposed method, particularly for quite low  ...  Acknowledgement This work was partially supported by Grant-in-Aid for Scientific Research (S) 21220003 and "R&D Program for Implementation of Anti-Crime and Anti-Terrorism Technologies for a Safe and Secure  ... 
doi:10.1109/cvpr.2012.6247844 dblp:conf/cvpr/AkaeMMY12 fatcat:co5axvqlzvcbtpm4vqhylsjysu

Cross View Gait Recognition Using Correlation Strength

Khalid Bashir, Tao Xiang, Shaogang Gong
2010 Procedings of the British Machine Vision Conference 2010  
To measure the similarity of gait sequences captured at different view angles, we model the correlation of gait sequences from different views using Canonical Correlation Analysis (CCA) and use the correlation  ...  We formulate a Gaussian Process (GP) classification framework to estimate the view angle of each probe gait sequence.  ...  Approaches in the second category either use a single camera and assume the subjects to be far away from the camera and perform a view synthesis for an arbitrary view using planar imaging geometry [5,  ... 
doi:10.5244/c.24.109 dblp:conf/bmvc/BashirXG10 fatcat:4xkxtmzdxfgkbblh4tqiijyq74

Bilinear Active Appearance Models

Jose Gonzalez-Mora, Fernando De la Torre, Rajesh Murthi, Nicolas Guil, Emilio L. Zapata
2007 2007 IEEE 11th International Conference on Computer Vision  
In particular, Active Appearance Models (AAMs) have been successfully used for face tracking, synthesis and recognition, and they are one of the state-of-the-art approaches due to its efficiency and representational  ...  This paper proposes Bilinear Active Appearance Models (BAAMs), an extension of AAMs, that effectively decouple changes due to pose and expression/identity.  ...  Abboud and Davoine [2, 3] use an asymmetric bilinear model to separate identity/expression parameters using AAMs [10] for face recognition and synthesis.  ... 
doi:10.1109/iccv.2007.4409185 dblp:conf/iccv/Gonzalez-MoraTMGZ07 fatcat:7hi6rtzmebagdhzl352qna7wyy

Human recognition in a video network

Bir Bhanu, Mingyue Ding, Bir Bhanu, Friedrich M. Wahl, Jonathan Roberts
2009 MIPPR 2009: Pattern Recognition and Computer Vision  
It addresses challenging questions for individual identification using gait and face at a distance and present new techniques and their comparison for robust identification.  ...  environments, and entertainment in public and private spaces.  ...  The contents of the information do not reflect the position or policy of the US Government.  ... 
doi:10.1117/12.850752 fatcat:lb5drxyvr5av5j43cud5zynt2e

Homeomorphic Manifold Analysis (HMA): Generalized separation of style and content on manifolds

Ahmed Elgammal, Chan-Su Lee
2013 Image and Vision Computing  
We show the application of the framework in synthesis, recognition, and tracking of certain human motions that follow this setting, such as gait and facial expressions.  ...  Given a set of topologically equivalent manifolds, the Homeomorphic Manifold Analysis (HMA) framework models the variation in their geometries in the space of functions that maps between a topologically-equivalent  ...  Facial Expression Synthesis and Recognition We learned and evaluated the model in Eq 6 using different datasets.  ... 
doi:10.1016/j.imavis.2012.12.003 fatcat:ejlsuqidezhmfpuasmbiajzhta

Multi-view face recognition by nonlinear tensor decomposition

Chunna Tian, Guoliang Fan, Xinbo Gao
2008 Pattern Recognition (ICPR), Proceedings of the International Conference on  
and the view-based PCA method.  ...  We use this technique to provide a generative face model that can deal with both the linearity and nonlinearity in multi-view face images.  ...  facial geometries, illumination, expressions and head poses etc [11] .  ... 
doi:10.1109/icpr.2008.4761195 dblp:conf/icpr/TianFG08 fatcat:cqk2xrjhfbbydgrbxhjtebewpi

"Dive into the Movie" Audience-Driven Immersive Experience in the Story

S. MORISHIMA
2008 IEICE transactions on information and systems  
Acknowledgments This study is supported by the Special Coordination Funds for Promoting Science and Technology of Ministry of Education, Sports, Science and Technology of Japan.  ...  I give special thanks to all members and supporters of Dive into the Movie project. References [1]  ...  Gait personality is modeled by PCA of captured joint angle motion vectors in generic full body skeleton model.  ... 
doi:10.1093/ietisy/e91-d.6.1594 fatcat:u7pfvqakibecxka6yha25hkxcy

Stylized synthesis of facial speech motions

Yuru Pei, Hongbin Zha
2007 Computer Animation and Virtual Worlds  
The visyllable modeling includes two parts: (1) A dynamic model for each kind of visyllable that is learnt based on a Gaussian Process Dynamical Model; (2) A multilinear model based unified mapping between  ...  In this paper, we address the fundamental issues regarding the stylized dynamic modeling of visyllables. The decomposable generalized model is learnt for the stylized motion synthesis.  ...  They would also like to thank the anonymous referees for the useful suggestions for improving this paper.  ... 
doi:10.1002/cav.186 fatcat:ttw5hxgrs5e7vpsmy6r4h5imcu
« Previous Showing results 1 — 15 out of 290 results