Results 1 - 10
of
12
Classification and recognition of dynamical models: the role of phase, independent components, kernels and optimal transport
- IEEE Trans. Pattern Anal. Mach. Intell
, 2007
"... Abstract—We address the problem of performing decision tasks and, in particular, classification and recognition in the space of dynamical models in order to compare time series of data. Motivated by the application of recognition of human motion in image sequences, we consider a class of models that ..."
Abstract
-
Cited by 29 (4 self)
- Add to MetaCart
(Show Context)
Abstract—We address the problem of performing decision tasks and, in particular, classification and recognition in the space of dynamical models in order to compare time series of data. Motivated by the application of recognition of human motion in image sequences, we consider a class of models that include linear dynamics, both stable and marginally stable (periodic), both minimum and nonminimum phases, driven by non-Gaussian processes. This requires extending existing learning and system identification algorithms to handle periodic modes and nonminimum-phase behavior while taking into account higher order statistics of the data. Once a model is identified, we define a kernel-based cord distance between models, which includes their dynamics, their initial conditions, and input distribution. This is made possible by a novel kernel defined between two arbitrary (non-Gaussian) distributions, which is computed by efficiently solving an optimal transport problem. 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 distance). However, our work can be applied more broadly where one needs to compare historical data while taking into account periodic trends, nonminimum-phase behavior, and non-Gaussian input distributions.
Bilinear modeling via Augmented Lagrange Multipliers (BALM)
- IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
, 2012
"... This paper presents a unified approach to solve different bilinear factorization problems in computer vision in the presence of missing data in the measurements. The problem is formulated as a constrained optimization where one of the factors must lie on a specific manifold. To achieve this, we intr ..."
Abstract
-
Cited by 12 (0 self)
- Add to MetaCart
This paper presents a unified approach to solve different bilinear factorization problems in computer vision in the presence of missing data in the measurements. The problem is formulated as a constrained optimization where one of the factors must lie on a specific manifold. To achieve this, we introduce an equivalent reformulation of the bilinear factorization problem that decouples the core bilinear aspect from the manifold specificity. We then tackle the resulting constrained optimization problem via Augmented Lagrange Multipliers. The strength and the novelty of our approach is that this framework can seamlessly handle different computer vision problems. The algorithm is such that only a projector onto the manifold constraint is needed. We present experiments and results for some popular factorization problems in computer vision such as rigid, non-rigid, and articulated Structure from Motion, photometric stereo, and 2D-3D non-rigid registration.
Action Recognition based on Homography Constraints
"... In this paper, we present a new approach for viewinvariant action recognition using constraints derived from the eigenvalues of planar homographies associated with triplets of body points. Unlike existing methods that study an action as a whole, or break it down into individual poses, we represent a ..."
Abstract
-
Cited by 3 (0 self)
- Add to MetaCart
(Show Context)
In this paper, we present a new approach for viewinvariant action recognition using constraints derived from the eigenvalues of planar homographies associated with triplets of body points. Unlike existing methods that study an action as a whole, or break it down into individual poses, we represent an action as a sequence of pose transitions. Using the fact that the homography induced by the motion of a triplet of body points in two identical pose transitions reduces to the special case of a homology, we exploit the equality of two of its eigenvalues to impose constraints on the similarity of the pose transitions between two subjects, observed by different perspective cameras and from different viewpoints. Experimental results show that our method can accurately identify human pose transitions and actions even when they include dynamic timeline maps, and are obtained from totally different viewpoints with different camera parameters. 1.
Trajectories normalization for viewpoint invariant gait recognition
- In ICPR
, 2008
"... This paper proposes a method to obtain frontoparallel (side-view) body part trajectories for a walk observed from an arbitrary view. The method is based on homography transformations computed for each gait half-cycle detected in the walk. Each homography maps the body part trajectories to a simulate ..."
Abstract
-
Cited by 2 (0 self)
- Add to MetaCart
(Show Context)
This paper proposes a method to obtain frontoparallel (side-view) body part trajectories for a walk observed from an arbitrary view. The method is based on homography transformations computed for each gait half-cycle detected in the walk. Each homography maps the body part trajectories to a simulated side view of the walk. The proposed method is stable as the resulting normalized trajectories are not influenced by missing or omitted parts of the raw trajectories. Experiments confirm that normalized trajectories are in agreement with the ones that would be obtained from a side view. 1.
Communication
"... Recognizing human actions are important in various real time applications. Review on human activity analysis is provided in three sections. The first section in this paper presents an overall classification to Human activity analysis from feature extraction to recognition systems. In the second sect ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
Recognizing human actions are important in various real time applications. Review on human activity analysis is provided in three sections. The first section in this paper presents an overall classification to Human activity analysis from feature extraction to recognition systems. In the second section a survey is included which provides technical information to activity analysis. Finally a brief description of databases which came across in survey is also included. The overall purpose of this paper is to provide a basic understanding to human activity analysis and to analyze the major challenge in human activity analysis.
Fisher Tensor Decomposition for Unconstrained Gait Recognition
"... Abstract. This paper proposes a simplified Tucker decomposition of a tensor model for gait recognition from dense local spatiotemporal (S/T) features extracted from gait video sequences. Unlike silhouettes, local S/T features have displayed state-of-art performances on challenging ac-tion recognitio ..."
Abstract
-
Cited by 1 (0 self)
- Add to MetaCart
(Show Context)
Abstract. This paper proposes a simplified Tucker decomposition of a tensor model for gait recognition from dense local spatiotemporal (S/T) features extracted from gait video sequences. Unlike silhouettes, local S/T features have displayed state-of-art performances on challenging ac-tion recognition testbeds, and have the potential to push gait ID towards real-world deployment. We adopt a Fisher representation of S/T features, rearranged as tensors. These tensors still contain redundant information, and are projected onto a lower dimensional space with tensor decompo-sition. The dimensions of the reduced tensor space can be automatically selected by keeping a proportion of the energy of the original tensor. Gait features can then be extracted from the reduced “core ” tensor, and ranked according to how relevant each feature is for classification. We validate our method on the benchmark USF/INIST gait data set, show-ing performances in line with the best reported results. 1
Towards View-Invariant Gait modeling: Computing View-Normalized Body Part Trajectories
"... This paper proposes an approach to compute view-normalized body part trajectories of pedestrians walking on potentially non-linear paths. The proposed approach finds applications in gait modeling, gait biometrics, and in medical gait analysis. Our approach uses the 2D trajectories of both feet and o ..."
Abstract
- Add to MetaCart
(Show Context)
This paper proposes an approach to compute view-normalized body part trajectories of pedestrians walking on potentially non-linear paths. The proposed approach finds applications in gait modeling, gait biometrics, and in medical gait analysis. Our approach uses the 2D trajectories of both feet and of the head extracted from the tracked silhouettes. On that basis, it computes the apparent walking (sagittal) planes for each detected gait half-cycle. A homography transformation is then computed for each walking plane to make it appear as if walking was observed from a fronto-parallel view. Finally, each homography is applied to head and feet trajectories over each corresponding gait half-cycle. View normalization makes head and feet trajectories appear as if seen from a fronto-parallel viewpoint, which is assumed to be optimal for gait modeling purposes. The proposed approach is fully automatic as it requires neither manual initialization nor camera calibration. An extensive experimental evaluation of the proposed approach confirms the validity of the normalization process.
Computing and Evaluating View-normalized Body Part Trajectories
"... This paper proposes an approach to compute and evaluate view-normalized body part trajectories of pedestrians from monocular video sequences. The proposed approach uses the 2D trajectories of both feet and of the head extracted from the tracked silhouettes. On that basis, it segments the walking tra ..."
Abstract
- Add to MetaCart
(Show Context)
This paper proposes an approach to compute and evaluate view-normalized body part trajectories of pedestrians from monocular video sequences. The proposed approach uses the 2D trajectories of both feet and of the head extracted from the tracked silhouettes. On that basis, it segments the walking trajectory into piecewise linear segments. Finally, a normalization process is applied to head and feet trajectories over each obtained straight walking segment. View normalization makes head and feet trajectories appear as if seen from a fronto-parallel viewpoint. The latter is assumed to be optimal for gait modeling and identification purposes. The proposed approach is fully automatic as it requires neither manual initialization nor camera calibration. An extensive experimental evaluation of the proposed approach confirms the validity of the normalization process. Key words: body parts trajectories, view-invariance, normalization, gait 1
Electronic Letters on Computer Vision and Image Analysis 8(1):15-26, 2009 Gait Identification Considering Body Tilt by Walking Direction Changes
, 2008
"... Gait identification has recently gained attention as a method of identifying individuals at a distance. Thought most of the previous works mainly treated straight-walk sequences for simplicity, curved-walk sequences should be also treated considering situations where a person walks along a curved pa ..."
Abstract
- Add to MetaCart
(Show Context)
Gait identification has recently gained attention as a method of identifying individuals at a distance. Thought most of the previous works mainly treated straight-walk sequences for simplicity, curved-walk sequences should be also treated considering situations where a person walks along a curved path or enters a building from a sidewalk. In such cases, person’s body sometimes tilts by centrifugal force when walking directions change, and this body tilt considerably degrades gait silhouette and identification performance, especially for widely-used appearance-based approaches. Therefore, we propose a method of body-tilted silhouette correction based on centrifugal force estimation from walking trajectories. Then, gait identification process including gait feature extraction in the frequency domain and learning of a View Transformation Model (VTM) follows the silhouette correction. Experiments of gait identification for circular-walk sequences demonstrate the effectiveness of the proposed method.
Learning manifolds of dynamical models for activity recognition
, 2010
"... In action, activity or identity recognition it is sometimes useful, rather than to extract some spatio-temporal features from the volumes representing motions, to explicitly encode their dynamics by means of dynamical systems, such as for instance hidden Markov models, nonlinear dynamical systems, o ..."
Abstract
- Add to MetaCart
(Show Context)
In action, activity or identity recognition it is sometimes useful, rather than to extract some spatio-temporal features from the volumes representing motions, to explicitly encode their dynamics by means of dynamical systems, such as for instance hidden Markov models, nonlinear dynamical systems, or hierarchical HMMs. Actions can then be classified by measuring distances in the appropriate space of models. However, using a fixed, arbitrary distance to classify dynamical models does not necessarily produce good classification results. The present proposal is concerned with a general differential-geometric framework for learning Riemannian metrics or distance functions for dynamical models, given a training set which can be either labeled or unlabeled. Given a training set of models, the optimal metric or distance function is selected among a family of pullback metrics induced by a parameterized automorphism of the space of models. The exploitation potential of the proposed methodology for action and activity recognition for human machine interaction, video indexing and summarization, is enormous.
