A soft-biometrics dataset for person tracking and re-identification

Schumann, Arne; Monari, Eduardo

doi:10.1109/avss.2014.6918667

Cited by 18 publications

(12 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In turn, this distrust could serve as a disincentive for organizations to implement this type of threat-avoidance solution. Other researchers have described methods of using disparate data sources along with security system data to reidentify otherwise anonymous individuals (Salvagnini, Bazzani, Cristani, & Murino, 2013;Schumann & Monari, 2014). In some cases, such as in analyzing keystrokes as a behavioral biometric, collecting the required security system data may lead to privacy risks if exposed (Sun & Upadhyaya, 2015).…”

Section: Distrust Propensitymentioning

confidence: 99%

Refining Technology Threat Avoidance Theory

Carpenter¹,

Young²,

Barrett³

et al. 2019

CAIS

View full text Add to dashboard Cite

Understanding individual threat avoidance motivation and behavior is a critical component in designing effective cyber security solutions for both users and organizations. Technology threat avoidance theory (TTAT) asserts that individuals' perceptions regarding their susceptibility to and the resulting severity of technology threats influence their awareness of the threats, which, in turn, influences their motivation and behavior to avoid them. While TTAT provides cogently and logically explains individuals' technology threat motivations and behaviors, empirical tests have produced equivocal results particularly in terms of the influence of susceptibility and severity on threat perceptions. Due to these inconsistencies in the threat calculus involving susceptibility, severity, and threat, we need more work to improve and understand individual threat motivations. Additionally, TTAT does not account for individual differences such as risk propensity, distrust propensity, and impulsivity that have been shown to affect cyber security behavior. To address these gaps, we present an empirical assessment of a refined TTAT model, which includes individual differences and models the influence of susceptibility on threat perceptions as partially mediated by severity. Results indicate that, while perceived susceptibility is a significant predictor of threat perceptions, severity perceptions partially mediates its effect. Our results also support the inclusion of risk propensity and distrust propensity in the TTAT model as personal characteristics that significantly affect overall threat perceptions.

show abstract

Section: Distrust Propensitymentioning

confidence: 99%

Refining Technology Threat Avoidance Theory

Carpenter¹,

Young²,

Barrett³

et al. 2019

CAIS

View full text Add to dashboard Cite

show abstract

“…For this development, we discard methods based on motion detection through prediction and other similar techniques 7,8,9,10 , inasmuch as the PTZ (pan-tilt-zoom) Dome cameras of the National Police of Colombia changes abruptly the pan, tilt and zoom. Because of that the backgrounds of the images change drastically making useless several techniques of video analysis.…”

Section: Street Theft Detection Using Deep Learningmentioning

confidence: 99%

Reduced computational cost prototype for street theft detection based on depth decrement in Convolutional Neural Network. Application to Command and Control Information Systems (C2IS) in the National Police of Colombia.

Suarez-Paez¹,

Salcedo-Gonzalez²,

Esteve³

et al. 2018

IJCIS

View full text Add to dashboard Cite

This paper shows the implementation of a prototype of street theft detector using the deep learning technique R-CNN (Region-Based Convolutional Network), applied in the Command and Control Information System (C2IS) of National Police of Colombia, the prototype is implemented using three models of CNN (Convolutional Neural Network), AlexNet, VGG16 and VGG19 comparing their computational cost measuring the image processing time, according to the complexity (depth) of each model. Finally, we conclude which model has the lowest computational cost and is more useful for the case of the National Police of Colombia.

show abstract

“…Other possible scenarios include using biometric samples to link data in ways not previously anticipated (Woodward, 1997). For example, an organization could use access door biometric login records in conjunction with softbiometric camera data such as clothing characteristics or hair style to re-identify an otherwise anonymous person walking through publicly accessible areas of the organization (Bazzani, Cristani, & Murino, 2013;Shumann & Monari, 2014). Although the finger pattern recognition algorithms used by current devices do not record actual fingerprints many individuals perceive these devices as storing the actual fingerprint rather than a summary of pattern data.…”

Section: Employee Distrustmentioning

confidence: 98%