Performance evaluation of crowd image analysis using the PETS2009 dataset

This paper presents the PETS2009 outdoor crowd image analysis surveillance dataset and the performance evaluation of people counting, detection and tracking results using the dataset submitted to five IEEE Performance Evaluation of Tracking and Surveillance (PETS) workshops. The evaluation was carried out using well established metrics developed in the Video Analysis and Content Extraction (VACE) programme and the CLassification of Events, Activities, and Relationships (CLEAR) consortium. The comparative evaluation highlights the detection and tracking performance of the authors’ systems in areas such as precision, accuracy and robustness and provides a brief analysis of the metrics themselves to provide further insights into the performance of the authors’ systems.

Enabling actor model for crowd sensing and IoT

The cloud is playing a very important role in wireless sensor network, crowd sensing and IoT data collection and processing. However, current cloud solutions lack of some features that hamper the innovation a number of other new services. We propose a cloud solution that provides these missing features as multi-cloud and device multi-tenancy relying in a whole different fully distributed paradigm, the actor model.

Proxy re-encryption schemes for IoT and crowd sensing

IoT, crowd sensing and smart cities will be a traffic challenge. New communication paradigms as asynchronous messaging carry and forward, scheduled delivery and temporary storage will be needed to manage network resources dynamically. Since traditional end to end security will require keeping security associations among devices for a long time draining valuable resources, we propose and evaluate the use of proxy re-encryption protocols in these scenarios as a solution for reliable and flexible security.

Robust background model for pixel based people counting using a single uncalibrated camera

Several pixel-based people counting methods have been developed over the years. Among these the product of scale-weighted pixel sums and a linear correlation coefficient is a popular people counting approach. However most approaches have paid little attention to resolving the true background and instead take all foreground pixels into account. With large crowds moving at varying speeds and with the presence of other moving objects such as vehicles this approach is prone to problems. In this paper we present a method which concentrates on determining the true-foreground, i.e. human-image pixels only. To do this we have proposed, implemented and comparatively evaluated a human detection layer to make people counting more robust in the presence of noise and lack of empty background sequences. We show the effect of combining human detection with a pixel-map based algorithm to i) count only human-classified pixels and ii) prevent foreground pixels belonging to humans from being absorbed into the background model. We evaluate the performance of this approach on the PETS 2009 dataset using various configurations of the proposed methods. Our evaluation demonstrates that the basic benchmark method we implemented can achieve an accuracy of up to 87% on sequence ¿S1.L1 13-57 View 001¿ and our proposed approach can achieve up to 82% on sequence ¿S1.L3 14-33 View 001¿ where the crowd stops and the benchmark accuracy falls to 64%.

PETS2009: dataset and challenge

This paper describes the crowd image analysis challenge that forms part of the PETS 2009 workshop. The aim of this challenge is to use new or existing systems for i) crowd count and density estimation, ii) tracking of individual(s) within a crowd, and iii) detection of separate flows and specific crowd events, in a real-world environment. The dataset scenarios were filmed from multiple cameras and involve multiple actors.

An overview of the PETS 2009 challenge

This paper describes the crowd image analysis challenge that forms part of the PETS 2009 workshop. The aim of this challenge is to use new or existing systems for i) crowd count and density estimation, ii) tracking of individual(s) within a crowd, and iii) detection of separate flows and specific crowd events, in a real-world environment. The dataset scenarios were filmed from multiple cameras and involve multiple actors.

Overall evaluation of the PETS2009 results

This paper presents the results of the crowd image analysis challenge, as part of the PETS 2009 workshop. The evaluation is carried out using a selection of the metrics available in the Video Analysis and Content Extraction (VACE) program and the CLassification of Events, Activities, and Relationships (CLEAR) consortium. The evaluation highlights the strengths of the authors’ systems in areas such as precision, accuracy and robustness.

PETS2009 and Winter-PETS 2009 results: a combined evaluation

This paper presents the results of the crowd image analysis challenge of the Winter PETS 2009 workshop. The evaluation is carried out using a selection of the metrics developed in the Video Analysis and Content Extraction (VACE) program and the CLassification of Events, Activities, and Relationships (CLEAR) consortium [13]. The evaluation highlights the detection and tracking performance of the authors’systems in areas such as precision, accuracy and robustness. The performance is also compared to the PETS 2009 submitted results.

PETS2010 and PETS2009 evaluation of results using individual ground truthed single views

This paper presents the results of the crowd image analysis challenge of the PETS2010 workshop. The evaluation was carried out using a selection of the metrics developed in the Video Analysis and Content Extraction (VACE) program and the CLassification of Events, Activities, and Relationships (CLEAR) consortium. The PETS 2010 evaluation was performed using new ground truthing create from each independant two dimensional view. In addition, the performance of the submissions to the PETS 2009 and Winter-PETS 2009 were evaluated and included in the results. The evaluation highlights the detection and tracking performance of the authors’ systems in areas such as precision, accuracy and robustness.

Visual search for basic emotional expressions in autism; impaired processing of anger, fear and sadness, but a typical happy face advantage

Facial expression recognition was investigated in 20 males with high functioning autism (HFA) or Asperger syndrome (AS), compared to typically developing individuals matched for chronological age (TD CA group) and verbal and non-verbal ability (TD V/NV group). This was the first study to employ a visual search, “face in the crowd” paradigm with a HFA/AS group, which explored responses to numerous facial expressions using real-face stimuli. Results showed slower response times for processing fear, anger and sad expressions in the HFA/AS group, relative to the TD CA group, but not the TD V/NV group. Reponses to happy, disgust and surprise expressions showed no group differences. Results are discussed with reference to the amygdala theory of autism.

A comparative review of skill assessment: performance, prediction and profiling

The assessment of chess players is an increasingly attractive opportunity and an unfortunate necessity. The chess community needs to limit potential reputational damage by inhibiting cheating and unjustified accusations of cheating: there has been a recent rise in both. A number of counter-intuitive discoveries have been made by benchmarking the intrinsic merit of players’ moves: these call for further investigation. Is Capablanca actually, objectively the most accurate World Champion? Has ELO rating inflation not taken place? Stimulated by FIDE/ACP, we revisit the fundamentals of the subject to advance a framework suitable for improved standards of computational experiment and more precise results. Other domains look to chess as the demonstrator of good practice, including the rating of professionals making high-value decisions under pressure, personnel evaluation by Multichoice Assessment and the organization of crowd-sourcing in citizen science projects. The ‘3P’ themes of performance, prediction and profiling pervade all these domains.

Task relevance of emotional information affects anxiety-linked attention bias in visual search

Task relevance affects emotional attention in healthy individuals. Here, we investigate whether the association between anxiety and attention bias is affected by the task relevance of emotion during an attention task. Participants completed two visual search tasks. In the emotion-irrelevant task, participants were asked to indicate whether a discrepant face in a crowd of neutral, middle-aged faces was old or young. Irrelevant to the task, target faces displayed angry, happy, or neutral expressions. In the emotion-relevant task, participants were asked to indicate whether a discrepant face in a crowd of middle-aged neutral faces was happy or angry (target faces also varied in age). Trait anxiety was not associated with attention in the emotion-relevant task. However, in the emotion-irrelevant task, trait anxiety was associated with a bias for angry over happy faces. These findings demonstrate that the task relevance of emotional information affects conclusions about the presence of an anxiety-linked attention bias.

Amor intellectualis diaboli

An essay on love and liberty in the writings of Gillian Rose, Marquis de Sade, Max Horkheimer and Theodor W. Adorno, written in response to the following provocation: "Encore un effort. A banderole publicitaire carries the breathless descriptions of the new fashions for 1968, when anything goes and details place the accent on this or that part of the body and its adornment: a pair of shoes that come off in a struggle, for example, the heel of one snapped off; a striking checked shirt, with two buttons undone; a light-coloured trench coat (perfect for a May day); a blouson-style jacket that allows easy freedom of movement; place casual slacks worn with an ankle boot. Beauty is in the streets as fashion becomes democratic (or so say the houses of haute couture), while the philosopher of the boudoir extorts us once again to make an effort if we wish to be republicans. Here, to an assembled crowd of sensitive men and women, which petit-maitre or dangerous man of principles would suggest that the only moral system to reinforce political revolution is that of libertinage, the revenge of nature's course against the aberrations of society?"

Spatial patterns and environmental constraints on ecosystem services at a catchment scale

Improved understanding and prediction of the fundamental environmental controls on ecosystem service supply across the landscape will help to inform decisions made by policy makers and land-water managers. To evaluate this issue for a local catchment case study, we explored metrics and spatial patterns of service supply for water quality regulation, agriculture production, carbon storage, and biodiversity for the Macronutrient Conwy catchment. Methods included using ecosystem models such as LUCI and JULES, integration of national scale field survey datasets, earth observation products and plant trait databases, to produce finely resolved maps of species richness and primary production. Analyses were done with both 1x1 km gridded and subcatchment data. A common single gradient characterised catchment scale ecosystem services supply with agricultural production and carbon storage at opposing ends of the gradient as reported for a national-scale assessment. Species diversity was positively related to production due to the below national average productivity levels in the Conwy combined with the unimodal relationship between biodiversity and productivity at the national scale. In contrast to the national scale assessment, a strong reduction in water quality as production increased was observed in these low productive systems. Various soil variables were tested for their predictive power of ecosystem service supply. Soil carbon, nitrogen, their ratio and soil pH all had double the power of rainfall and altitude, each explaining around 45% of variation but soil pH is proposed as a potential metric for ecosystem service supply potential as it is a simple and practical metric which can be carried out in the field with crowd-sourcing technologies now available. The study emphasises the importance of considering multiple ecosystem services together due to the complexity of covariation at local and national scales, and the benefits of exploiting a wide range of metrics for each service to enhance data robustness.