Human action categorization using conditional random field

Automatic human action recognition has been a challenging issue in the field of machine vision. Some high-level features such as SIFT, although with promising performance for action recognition, are computationally complex to some extent. To deal with this problem, we construct the features based on the Distance Transform of body contours, which is relatively simple and computationally efficient, to represent human action in the video. After extracting the features from videos, we adopt the Conditional Random Field for modeling the temporal action sequences. The proposed method is tested with an available standard dataset. We also testify the robustness of our method on various realistic conditions, such as body occlusion or intersection.

Human action recognition based on 3D SIFT and LDA model

How to recognize human action from videos captured by modern cameras efficiently and effectively is a challenge in real applications. Traditional methods which need professional analysts are facing a bottleneck because of their shortcomings. To cope with the disadvantage, methods based on computer vision techniques, without or with only a few human interventions, have been proposed to analyse human actions in videos automatically. This paper provides a method combining the three dimensional Scale Invariant Feature Transform (SIFT) detector and the Latent Dirichlet Allocation (LDA) model for human motion analysis. To represent videos effectively and robustly, we extract the 3D SIFT descriptor around each interest point, which is sampled densely from 3D Space-time video volumes. After obtaining the representation of each video frame, the LDA model is adopted to discover the underlying structure-the categorization of human actions in the collection of videos. Public available standard datasets are used to test our method. The concluding part discusses the research challenges and future directions.

Human action recognition based on pyramid histogram of oriented gradients

Human action recognition has been attracted lots of interest from computer vision researchers due to its various promising applications. In this paper, we employ Pyramid Histogram of Orientation Gradient (PHOG) to characterize human figures for action recognition. Comparing to silhouette-based features, the PHOG descriptor does not require extraction of human silhouettes or contours. Two state-space models, i.e.; Hidden Markov Model (HMM) and Conditional Random Field (CRF), are adopted to model the dynamic human movement. The proposed PHOG descriptor and the state-space models with respect to different parameters are tested using a standard dataset. We also testify the robustness of the method with respect to various unconstrained conditions and viewpoints. Promising experimental result demonstrates the effectiveness and robustness of our proposed method.

Human action segmentation via controlled use of missing data in HMMs

Segmentation of individual actions from a stream of human motion is an open problem in computer vision. This paper approaches the problem of segmenting higher-level activities into their component sub-actions using Hidden Markov Models modified to handle missing data in the observation vector. By controlling the use of missing data, action labels can be inferred from the observation vector during inferencing, thus performing segmentation and classification simultaneously. The approach is able to segment both prominent and subtle actions, even when subtle actions are grouped together. The advantage of this method over sliding windows and Viterbi state sequence interrogation is that segmentation is performed as a trainable task, and the temporal relationship between actions is encoded in the model and used as evidence for action labelling.

Visual perception of movement kinematics and the acquisition of "action prototypes"

Recognizing a class of movements as belonging to a "nominal" action category, such as walking, running, or throwing, is a fundamental human ability. Three experiments were undertaken to test the hypothesis that common ("prototypical") features of moving displays could be learned by observation. Participants viewed moving stick-figure displays resembling forearm flexion movements in the saggital plane. Four displays (presentation displays) were first presented in which one or more movement dimensions were combined with 2 respective cues: direction (up, down), speed (fast, slow), and extent (long, short). Eight test displays were then shown, and the observer indicated whether each test display was like or unlike those previously seen. The results showed that without corrective feedback, a single cue (e.g., up or down) could be correctly recognized, on average, with the proportion correct between .66 and .87. When two cues were manipulated (e.g., up and slow), recognition accuracy remained high, ranging between .72 and .89. Three-cue displays were also easily identified. These results provide the first empirical demonstration of action-prototype learning for categories of human action and show how apparently complex kinematic patterns can be categorized in terms of common features or cues. It was also shown that probability of correct recognition of kinematic properties was reduced when the set of 4 presentation displays were more variable with respect to their shared kinematic property, such as speed or amplitude. Finally, while not conclusive, the results (from 2 of the 3 experiments) did suggest that similarity (or "likeness") with respect to a common kinematic property (or properties) is more easily recognized than dissimilarity.

Action research in higher education : the advancement of university learning and teaching

This thesis aims to contribute to the improvement and advancement of university learning, teaching, and staff development; to integrate educational theory and the practice of university teaching; and to contribute to the establishment of a new, emerging paradigm in higher education. The strategy towards achieving these aims comprises (1) an alternative research methodology in the interpretive, non-positivist paradigm; (2) an integrated framework drawing on a variety of previously unrelated theories to form an alternative model of university education; and (3) reference to the dialectical relationship between educational theory and teaching practice and their integration through action research in higher education. The thesis is not so much a critique of the traditional paradigm and of existing functionalist-structuralist approaches to higher education, but more a development and clarification of an alternative, dialectical, human action approach to higher education. The original contribution of this thesis to the theory and practice of higher education lies in the development (1) of an alternative model of university education based on an integration of previously unrelated domains of theory; (2) of a theoretical model of professional development as action research (the CRASP Model: Critical attitude, Research into teaching, Accountability, Self-evaluation, Professionalism); and (3) of action research projects in higher education. Action research is research by the university teachers themselves into their teaching practice, i.e. into problems of the curriculum and student learning. The case studies included in and appended to this thesis show that in one educational setting at least it was possible to improve and advance university learning and teaching through action research. The evidence for this advancement is provided in a number of previously published case studies compiled in the Appendix.

Classifying human actions using an incomplete real-time pose skeleton

Currently, most human action recognition systems are trained with feature sets that have no missing data. Unfortunately, the use of human pose estimation models to provide more descriptive features also entails an increased sensitivity to occlusions, meaning that incomplete feature information will be unavoidable for realistic scenarios. To address this, our approach is to shift the responsibility for dealing with occluded pose data away from the pose estimator and onto the action classifier. This allows the use of a simple, real-time pose estimation (stick-figure) that does not estimate the positions of limbs it cannot find quickly. The system tracks people via background subtraction and extracts the (possibly incomplete) pose skeleton from their silhouette. Hidden Markov Models modified to handle missing data are then used to successfully classify several human actions using the incomplete pose features.

A syntactic two-component encoding model for the trajectories of human actions

Human actions have been widely studied for their potential application in various areas such as sports, pervasive patient monitoring, and rehabilitation. However, challenges still persist pertaining to determining the most useful ways to describe human actions at the sensor, then limb and complete action levels of representation and deriving important relations between these levels each involving their own atomic components. In this paper, we report on a motion encoder developed for the sensor level based on the need to distinguish between the shape of the sensor's trajectory and its temporal characteristics during execution. This distinction is critical as it provides a different encoding scheme than the usual velocity and acceleration measures which confound these two attributes of any motion. At the same time, we eliminate noise from sensors by comparing temporal and spatial indexing schemes and a number of optimal filtering models for robust encoding. Results demonstrate the benefits of spatial indexing and separating the shape and dynamics of a motion, as well as its ability to decompose complex motions into several atomic ones. Finally, we discuss how this specific type of sensor encoder bears on the derivation of limb and complete action descriptions.

Toddler techies: a study of young children`s interaction with computers

This article describes an ethnographic study of children's behavioural interaction with multimedia within a familiar context. The rationale for such a study was to provide data and evaluation of the capabilities of young children in an expressly modified multimedia environment and to determine the usefulness of employing technology as an adjunct to young children's play. However, hermeneutic and interpretativist concerns for the study of human action and social practice in the use of technology also informs both the structural, procedural, and evaluative management of the study. Using customised children's software, observation focused on time spent using the computer, the attitude toward the computer, the reaction to the interface, their use and adaptation of the mouse, and adult interventions. Significantly, the results differ appreciably from previous research and possible grounds for this variation is explored.

Switching on the chronometer of history : communism, postcommunism and heritage

The destruction of monuments accompanying the fall of Communism ignited debates about preservation of manifestations of a hated regime. While heritage professionals called for their preservation as ‘historical documents’, many monuments were destroyed or removed. Yampolsky sees anti-Communist iconoclasm as a rejection of the totalitarianism of time embodied in Communist monuments. These ‘intentional monuments’ were intended to ‘negate the march of time and oppose to it the permanence of human action’. They demonstrated the alleged end of history in a classless utopia.

Iconoclastic acts against these monuments involved the crossing of ‘the invisible boundaries of the sacral zone surrounding monuments, switching on the chronometer of history’. In doing so, iconoclasts provide the conditions for reassertion of heritage practices: heritage requires a sense of the flow of time, a difference between past, present and future.

Having restarted the chronometer of history, a society is forced to assess where it stands in relation to its past. Will it continue on a path of ‘wilful forgetting’, or seek to confront the past? The danger of wilful forgetting is the creation of nostalgia. Alternatively, preservation of places of memory helps processing of the past required for movement into the future. ‘One need only consider the way in which Berliners tore down the hated Berlin Wall in the aftermath of 1989’, Fulbrook writes, ‘to understand the desire to rid the landscape of a hated excrescence, a symbol of a rejected political past. But…for those who come after, the effort of historical imagination is all the greater for lack of a topography of experience’.

Heritage preservation can produce a ‘topography of experience’, through which the experience of Communism is examined. Reassertion of a humanistic historical time through heritage practices reveals the arrogant futility of utopian projects seeking to bring history to an end.

An integrated theory of sexual offending

This paper represents a first attempt to provide an integrated framework to explain the onset, development, and maintenance of sexual offending. According to the Integrated Theory of Sexual Offending (ITSO), sexual abuse occurs as a consequence of a number of interacting causal variables. We examine the factors that affect brain development (evolution, genetic variations and neurobiology) and ecological factors (social and cultural environment, personal circumstances, physical environment) and discuss how they impact upon core neuropsychological functions underpinning human action. The ITSO then explains how clinical symptoms arise from the interaction between these neurological systems and ecological factors. The capacity of the ITSO to incorporate competing theories of sexual offending is considered, and we end the paper by critically evaluating its usefulness in stimulating research and further theory development.

Improving science uptake in coastal zone management : principles for science engagement and their application in south-eastern Tasmania

Sustainable management of the coastal zone represents a considerable challenge to Australian society. This challenge is rooted in the complexity of the biophysical and sociocultural characteristics of coastal areas, including uncertainty about system characteristics and processes, and the diversity of stakeholders, their interests, values and perspectives, and the jurisdictions involved in coastal governance and management. Given this complexity of coastal zone management, scientific and other forms of knowledge can affect decision-making and human action in diverse ways, which will often depend on the ability of scientists to engage effectively with relevant stakeholders.

Synergistic antiproliferative action of the Flavonols Quercetin and Kaempferol in cultured human cancer cell lines

The consumption of vegetables containing the flavonols quercetin and kaempferol reduces the risk of cancer. We utilized human gut (HuTu-80 and Caco-2) and breast cancer cells (PMC42) to show the synergistic effect of quercetin and kaempferol in reducing cell proliferation. A trend in reduction of total cell counts was seen following a single exposure, a 4-day exposure or a 14-day exposure to quercetin and kaempferol. Combined treatments with quercetin and kaempferol were more effective than the additive effects of each flavonol. The reduction in cell proliferation was associated with decreased expression of nuclear proliferation antigen Ki67 and decreased total protein levels in treated cells relative to controls. In conclusion, the synergistic antiproliferative effect of quercetin and kaempferol demonstrated in cultured human cells has broad implications for understanding the influence of dietary nutrients in vivo, where anticancer effects may be a result of nutrients which act in concert.