Activity modelling in crowded environments : a soft decision approach

Probabilistic topic models have recently been used for activity analysis in video processing, due to their strong capacity to model both local activities and interactions in crowded scenes. In those applications, a video sequence is divided into a collection of uniform non-overlaping video clips, and the high dimensional continuous inputs are quantized into a bag of discrete visual words. The hard division of video clips, and hard assignment of visual words leads to problems when an activity is split over multiple clips, or the most appropriate visual word for quantization is unclear. In this paper, we propose a novel algorithm, which makes use of a soft histogram technique to compensate for the loss of information in the quantization process; and a soft cut technique in the temporal domain to overcome problems caused by separating an activity into two video clips. In the detection process, we also apply a soft decision strategy to detect unusual events.We show that the proposed soft decision approach outperforms its hard decision counterpart in both local and global activity modelling.

Real-time video event detection in crowded scenes using MPEG derived features : a multiple instance learning approach

This paper presents an investigation into event detection in crowded scenes, where the event of interest co-occurs with other activities and only binary labels at the clip level are available. The proposed approach incorporates a fast feature descriptor from the MPEG domain, and a novel multiple instance learning (MIL) algorithm using sparse approximation and random sensing. MPEG motion vectors are used to build particle trajectories that represent the motion of objects in uniform video clips, and the MPEG DCT coefficients are used to compute a foreground map to remove background particles. Trajectories are transformed into the Fourier domain, and the Fourier representations are quantized into visual words using the K-Means algorithm. The proposed MIL algorithm models the scene as a linear combination of independent events, where each event is a distribution of visual words. Experimental results show that the proposed approaches achieve promising results for event detection compared to the state-of-the-art.

Automatic classification of human epithelial type 2 cell indirect immunofluorescence images using cell pyramid matching

This paper describes a novel system for automatic classification of images obtained from Anti-Nuclear Antibody (ANA) pathology tests on Human Epithelial type 2 (HEp-2) cells using the Indirect Immunofluorescence (IIF) protocol. The IIF protocol on HEp-2 cells has been the hallmark method to identify the presence of ANAs, due to its high sensitivity and the large range of antigens that can be detected. However, it suffers from numerous shortcomings, such as being subjective as well as time and labour intensive. Computer Aided Diagnostic (CAD) systems have been developed to address these problems, which automatically classify a HEp-2 cell image into one of its known patterns (eg. speckled, homogeneous). Most of the existing CAD systems use handpicked features to represent a HEp-2 cell image, which may only work in limited scenarios. We propose a novel automatic cell image classification method termed Cell Pyramid Matching (CPM), which is comprised of regional histograms of visual words coupled with the Multiple Kernel Learning framework. We present a study of several variations of generating histograms and show the efficacy of the system on two publicly available datasets: the ICPR HEp-2 cell classification contest dataset and the SNPHEp-2 dataset.

The role of the posterior superior temporal sulcus in audiovisual processing

In this study we investigate previous claims that a region in the left posterior superior temporal sulcus (pSTS) is more activated by audiovisual than unimodal processing. First, we compare audiovisual to visual-visual and auditory-auditory conceptual matching using auditory or visual object names that are paired with pictures of objects or their environmental sounds. Second, we compare congruent and incongruent audiovisual trials when presentation is simultaneous or sequential. Third, we compare audiovisual stimuli that are either verbal (auditory and visual words) or nonverbal (pictures of objects and their associated sounds). The results demonstrate that, when task, attention, and stimuli are controlled, pSTS activation for audiovisual conceptual matching is 1) identical to that observed for intramodal conceptual matching, 2) greater for incongruent than congruent trials when auditory and visual stimuli are simultaneously presented, and 3) identical for verbal and nonverbal stimuli. These results are not consistent with previous claims that pSTS activation reflects the active formation of an integrated audiovisual representation. After a discussion of the stimulus and task factors that modulate activation, we conclude that, when stimulus input, task, and attention are controlled, pSTS is part of a distributed set of regions involved in conceptual matching, irrespective of whether the stimuli are audiovisual, auditory-auditory or visual-visual.

SAIVT-ADMRG @ MediaEval 2014 social event detection

This paper outlines the approach taken by the Speech, Audio, Image and Video Technologies laboratory, and the Applied Data Mining Research Group (SAIVT-ADMRG) in the 2014 MediaEval Social Event Detection (SED) task. We participated in the event based clustering subtask (subtask 1), and focused on investigating the incorporation of image features as another source of data to aid clustering. In particular, we developed a descriptor based around the use of super-pixel segmentation, that allows a low dimensional feature that incorporates both colour and texture information to be extracted and used within the popular bag-of-visual-words (BoVW) approach.

Supervised latent dirichlet allocation models for efficient activity representation

Local spatio-temporal features with a Bag-of-visual words model is a popular approach used in human action recognition. Bag-of-features methods suffer from several challenges such as extracting appropriate appearance and motion features from videos, converting extracted features appropriate for classification and designing a suitable classification framework. In this paper we address the problem of efficiently representing the extracted features for classification to improve the overall performance. We introduce two generative supervised topic models, maximum entropy discrimination LDA (MedLDA) and class- specific simplex LDA (css-LDA), to encode the raw features suitable for discriminative SVM based classification. Unsupervised LDA models disconnect topic discovery from the classification task, hence yield poor results compared to the baseline Bag-of-words framework. On the other hand supervised LDA techniques learn the topic structure by considering the class labels and improve the recognition accuracy significantly. MedLDA maximizes likelihood and within class margins using max-margin techniques and yields a sparse highly discriminative topic structure; while in css-LDA separate class specific topics are learned instead of common set of topics across the entire dataset. In our representation first topics are learned and then each video is represented as a topic proportion vector, i.e. it can be comparable to a histogram of topics. Finally SVM classification is done on the learned topic proportion vector. We demonstrate the efficiency of the above two representation techniques through the experiments carried out in two popular datasets. Experimental results demonstrate significantly improved performance compared to the baseline Bag-of-features framework which uses kmeans to construct histogram of words from the feature vectors.

A picture is worth a thousand words : an approach to learning about visuals

Students in the middle years encounter an increasing range of unfamiliar visuals. Visual literacy, the ability to encode and decode visuals and to think visually, is an expectation of all middle years curriculum areas and an expectation of NAPLAN literacy and numeracy tests. This article presents a multidisciplinary approach to teaching visual literacy that links the content of all learning areas and encourages students to transfer skills from familiar to unfamiliar contexts. It proposes a classification of visuals in six parts: one-dimensional; two-dimensional; map; shape; connection; and picture, based on the properties, rather than the purpose, of the visual. By placing a visual in one of these six categories, students learn to transfer the skills used to decode familiar visuals to unfamiliar cases in the same category. The article also discusses a range of other teaching strategies that can be used to complement this multi-disciplinary approach.

A picture is worth a thousand words : a cross-curricular approach to learning about visuals in STEM

Visuals are a central feature of STEM in all levels of education and many areas of employment. The wide variety of visuals that students are expected to master in STEM prevents an approach that aims to teach students about every type of visual that they may encounter. This paper proposes a pedagogy that can be applied across year levels and learning areas, allowing a school-wide, cross-curricular, approach to teaching about visual, that enhances learning in STEM and all other learning areas. Visuals are classified into six categories based on their properties, unlike traditional methods that classify visuals according to purpose. As visuals in the same category share common properties, students are able to transfer their knowledge from the familiar to unfamiliar in each category. The paper details the classification and proposes some strategies that can be can be incorporated into existing methods of teaching students about visuals in all learning areas. The approach may also assist students to see the connections between the different learning areas within and outside STEM.

The effect of prior visual information on recognition of speech and sounds

To identify and categorize complex stimuli such as familiar objects or speech, the human brain integrates information that is abstracted at multiple levels from its sensory inputs. Using cross-modal priming for spoken words and sounds, this functional magnetic resonance imaging study identified 3 distinct classes of visuoauditory incongruency effects: visuoauditory incongruency effects were selective for 1) spoken words in the left superior temporal sulcus (STS), 2) environmental sounds in the left angular gyrus (AG), and 3) both words and sounds in the lateral and medial prefrontal cortices (IFS/mPFC). From a cognitive perspective, these incongruency effects suggest that prior visual information influences the neural processes underlying speech and sound recognition at multiple levels, with the STS being involved in phonological, AG in semantic, and mPFC/IFS in higher conceptual processing. In terms of neural mechanisms, effective connectivity analyses (dynamic causal modeling) suggest that these incongruency effects may emerge via greater bottom-up effects from early auditory regions to intermediate multisensory integration areas (i.e., STS and AG). This is consistent with a predictive coding perspective on hierarchical Bayesian inference in the cortex where the domain of the prediction error (phonological vs. semantic) determines its regional expression (middle temporal gyrus/STS vs. AG/intraparietal sulcus).