Chaotic Invariants for Human Action Recognition

Human action recognition is one of the most promising research areas at the moment. In this paper, we present an entropy based effective approach to recognize human activities from video sequences. Our approach provides a new description directly from Motion History Image (MHI) which is an efficient real-time representation for human action. Local entropy of pixels in MHI is computed to characterize the texture of motion filed. It helps to match the moment-based feature statistically. In an evaluation of this on well established benchmark data sets, we achieve high recognition rates. Using the popular Weizmann data set, we achieve the best accuracy of 98.9% in a leave-one-out cross validation procedure. Using the well-known KTH data set, we evaluate the robustness of the proposed approach over different scenarios. As a result, this new method is computationally efficient, robust with respect to appearance variation, and straight forward to be applied as it builds itself on well established and understandable concepts.

Computer Vision and Image Understanding, 2005

Static and temporally varying 3D invariants are proposed for capturing the spatio-temporal dynamics of a general human action to enable its representation in a compact, view-invariant manner. Two variants of the representation are presented and studied: (1) a restricted-3D version, whose theory and implementation are simple and efficient but which can be applied only to a restricted class of human action, and (2) a full-3D version, whose theory and implementation are more complex but which can be applied to any general human action. A detailed analysis of the two representations is presented. We show why a straightforward implementation of the key ideas does not work well in the general case, and present strategies designed to overcome inherent weaknesses in the approach. What results is an approach for human action modeling and recognition that is not only invariant to viewpoint, but is also robust enough to handle different people, different speeds of action (and hence, frame rate) and minor variabilities in a given action, while encoding sufficient distinction among actions. Results on 2D projections of human motion capture and on manually segmented real image sequences demonstrate the effectiveness of the approach.

Statistical classification of actions in videos is mostly performed by extracting relevant features, particularly covariance features, from image frames and studying time series associated with temporal evolutions of these features. A natural mathematical representation of activity videos is in form of parameterized trajectories on the covariance manifold, i.e. the set of symmetric, positive-definite matrices (SPDMs). The variable execution-rates of actions implies variable parameterizations of the resulting trajectories, and complicates their classification. Since action classes are invariant to execution rates, one requires rate-invariant metrics for comparing trajectories. A recent paper represented trajectories using their transported square-root vector fields (TSRVFs), defined by parallel translating scaled-velocity vectors of trajectories to a reference tangent space on the manifold. To avoid arbitrariness of selecting the reference and to reduce distortion introduced during t...

Lecture Notes in Bioengineering, 2021

Human action recognition is a challenging area of research in the field of computer vision due to its complex analytical structure. Researches have been carried out in this field over the past few years. Now, as it has become possible to recognize human action successfully, the researchers are trying to make the recognition system real time in nature. But due to its complex video analysis, it has not been possible for the researchers to make it real time completely. Though some algorithms had been developed that will successfully recognize human action with the minimum delay possible in order to make the recognition system almost real time in nature. In this work, we have compared some of the successful human action recognition methodologies along with the recent trends to accomplish the task in negligible time delay.

PLoS ONE, 2012

Human and many other animals can detect, recognize, and classify natural actions in a very short time. How this is achieved by the visual system and how to make machines understand natural actions have been the focus of neurobiological studies and computational modeling in the last several decades. A key issue is what spatial-temporal features should be encoded and what the characteristics of their occurrences are in natural actions. Current global encoding schemes depend heavily on segmenting while local encoding schemes lack descriptive power. Here, we propose natural action structures, i.e., multi-size, multi-scale, spatial-temporal concatenations of local features, as the basic features for representing natural actions. In this concept, any action is a spatial-temporal concatenation of a set of natural action structures, which convey a full range of information about natural actions. We took several steps to extract these structures. First, we sampled a large number of sequences of patches at multiple spatial-temporal scales. Second, we performed independent component analysis on the patch sequences and classified the independent components into clusters. Finally, we compiled a large set of natural action structures, with each corresponding to a unique combination of the clusters at the selected spatial-temporal scales. To classify human actions, we used a set of informative natural action structures as inputs to two widely used models. We found that the natural action structures obtained here achieved a significantly better recognition performance than lowlevel features and that the performance was better than or comparable to the best current models. We also found that the classification performance with natural action structures as features was slightly affected by changes of scale and artificially added noise. We concluded that the natural action structures proposed here can be used as the basic encoding units of actions and may hold the key to natural action understanding.

British Journal of Applied Science & Technology, 2017

Despite their attractive properties of invariance, robustness and reliability, statistical motion descriptions from temporal templates have not apparently received the amount of attention they might deserve in the human action recognition literature. In this paper, we propose an innovative approach for action recognition, where a novel fuzzy representation based on temporal motion templates is developed to model human actions as time series of low-dimensional descriptors. An NB (Naïve Bayes) classifier is trained on these features for action classification. When tested on a realistic action dataset incorporating a large collection of video data, the results demonstrate that the approach is able to achieve a recognition rate of as high as 93.7%, while remaining tractable for real-time operation.

2015

The motion sequences of human actions have its own discriminating profile that can be represented as a spatiotemporal template like Motion History Image (MHI). A histogram is a popular statistic to present the underlying information in a template. In this paper a histogram oriented action recognition method is presented. In the proposed method, we use the Directional Motion History Images (DMHI), their corresponding Local Binary Pattern (LBP) images and the Motion Energy Image (MEI) as spatiotemporal template. The intensity histogram is then extracted from those images which are concatenated together to form the feature vector for action representation. A linear combination of the histograms taken from DMHIs and LBP images is used in the experiment. We evaluated the performance of the proposed method along with some variants of it using the renowned KTH action dataset and found higher accuracies. The obtained results justify the superiority of the proposed method compared to other a...

2014

This paper presents a human action recognition method using dynamic time warping and voting algorithms on 3D human skeletal models. In this method human actions, which are the combinations of multiple body pa rt movements, are described by feature matrices concer ning both spatial and temporal domains. The feature matrices are created based on the spatial selection of relative angles between body parts in time seri es. Then, action recognition is done by applying a classifier which is the combination of dynamic time warping ( DTW) and a voting algorithm to the feature matrices. Exp erimental results show that the performance of our action recognition method obtains high recognition accurac y at reliable computation speed and can be applied in real time human action recognition systems.

In this paper, we describe a simple strategy for mitigating variability in temporal data series by shifting focus onto long-term, frequency domain features that are less susceptible to variability. We apply this method to the human action recognition task and demonstrate how working in the frequency domain can yield good recognition features for commonly used optical flow and articulated pose features, which are highly sensitive to small differences in motion, viewpoint, dynamic backgrounds, occlusion and other sources of variability. We show how these frequency-based features can be used in combination with a simple forest classifier to achieve good and robust results on the popular KTH Actions dataset. * This work was partially supported by the DARPA Minds Eye program. The U.S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright annotation thereon. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of DARPA or the U.S. Government.

Vnu Journal of Science Computer Science and Communication Engineering, 2014

This paper presents a human action recognition method using dynamic time warping and voting algorithms on 3D human skeletal models. In this method human actions, which are the combinations of multiple body part movements, are described by feature matrices concerning both spatial and temporal domains. The feature matrices are created based on the spatial selection of relative angles between body parts in time series. Then, action recognition is done by applying a classifier which is the combination of dynamic time warping (DTW) and a voting algorithm to the feature matrices. Experimental results show that the performance of our action recognition method obtains high recognition accuracy at reliable computation speed and can be applied in real time human action recognition systems.