TU Berlin

Fachgebiet NachrichtenübertragungPrivacy Preserving Technologies

Inhalt des Dokuments

zur Navigation

Privacy Preserving Technologies

The ever increasing number of surveillance camera networks being deployed all over the world has resulted in a high interest in the development of algorithms to automatically analyze the video footage, but has also opened new questions as how to efficiently manage the vast amount of information generated and, more important, how to protect the privacy of the individuals being recorded in their daily life. An technical approach are the Privacy Preserving Technologies (PET) which are algorithm to filter personal privacy related information and/or adapt the protection level of personal privacy in videos.

Overview of the MediaEval 2014 Visual Privacy Task

Lupe

This paper presents an overview of the Visual Privacy Task (VPT) of MediaEval 2014, its objectives, related dataset, and evaluation approaches. Participants in this task were required to implement a privacy filter or a combination of filters to protect various personal information regions in video sequences as provided. The challenge was to achieve an adequate balance between the degree of privacy protection, intelligibility (how much useful information is retained post privacy filtering), and pleasantness (how minimal were the adverse effects of filtering on the appearance of the video frames). The submissions from the eight (8) teams who participated in this task were evaluated subjectively by surveillance experts, practitioners, data protection experts and by naïve viewers using a crowdsourcing approach.

Publication

Badii, A., Ebrahimi, T., Fedorczak, C., Korshunov, P., Piatrik, T., Eiselein, V., Ahmed, A. Overview of the MediaEval 2014 visual privacy taskMediaEval, 2014

 

 

MediaEval 2014 Visual Privacy Task: Reversible Scrambling on Foreground Masks

Lupe

This paper describes our participation in the Visual Privacy Task of MediaEval 2014, which aims to obscure human occurrence in image sequences. As a result the recorded person should be unrecognisable, but if needed the obscured areas can be recovered. We use an approach which models the background and pseudo-randomly scrambles pixels within disjunct foreground areas. This technique is reversible and preserves the colour characteristic of each area. So, colourbased approaches will still be able to automatically distinguish between differently dressed individuals. The evaluations of our results show that the privacy aspect got a high score in all three evaluation streams. The level of intelligibility and the pleasantness of our approach is below the average, since scrambling results in lower ‘aesthetic’ images.

Publication

Schmiedeke, S., Kelm, P., Goldmann, L., Sikora, T., TUB @ MediaEval 2014 Visual Privacy Task: Reversible Scrambling on Foreground MasksMediaEval, 2014

 

 

MediaEval 2013 Visual Privacy Task: Using Adaptive Edge Detection for Privacy in Surveillance Videos

Lupe

In this paper we present a system for preserving the privacy of individuals in a video surveillance scenario. While a person’s privacy should not be revealed to a viewer of the video without special needs, it is still important that the action in a scene as the semantic content of a video remain perceivable by a human observer. The proposed system uses edge detection and adaptive thresholding in order to estimate the persons’ silhouettes in a video scene and thus rendering most of their actions visible, while hiding sensitive personal information. In order to obtain a more complete contour around a person, an adaptive thresholding scheme using edge histograms is used as well as background subtraction which limits the edge extraction to foreground masks and thus avoids distraction of the viewer’s eyes to background structures.

Publication

Eiselein, V., Senst, T., Keller, I., Sikora, T. MediaEval 2013 Visual Privacy Task: Using Adaptive EdgeDetection for Privacy in Surveillance Videos, MediaEval, 2013

 

 

MediaEval 2013 Visual Privacy Task: Reversible Scrambling with colour-preservative Characteristic

Lupe

This paper describes our participation in the Visual Privacy Task of MediaEval 2013, which aims to obscure human occurrence in image sequences. As a result the recorded person should be unrecognisable. We use an approach which pseudo-randomly scrambles pixels within specified regions. This technique is reversible and preserves the colour characteristic of each region. So, colour-based approaches will still be able to automatically distinguish between differently dressed individuals. The evaluations of our results show that the privacy aspect got a very high score in both objective and subjective metrics. Our approach has a lack of intelligibility since it was measured by applying the Histogram of Oriented Gradients which might be fail on scrambled areas since edges are not preserved.

Publication

Schmiedeke, S., Kelm, P., Sikora, T. TUB @ MediaEval 2013 Visual Privacy Task: Reversible Scrambling with colour-preservative CharacteristicMediaEval, 2013

 

 

A decentralized Privacy-sensitive Video Surveillance Framework

Lupe

With the increasing spread of accurate and robust video surveillance, applications such as crowd monitoring, people counting and abnormal behavior recognition become ubiquitous.This leads to needs of interactive systems taking into account a high degree of interoperability as well as privacy protection concerns. In this paper we propose a framework based on the ONVIF specification to support the work of video operators while implementing a privacy-by-design concept.We use an OpenGL-based 3D model of the CCTV site where we display the results of the video analytics in an avatar-based manner and give an example application on mugging detection.To place the automatically detected scene information, such as people detections and events, an automatic camera calibration is used which effectively reduces the deployment effort.

Publication

Senst, T., Eiselein, V., Badii, A., Einig, M., Keller, I., Sikora, T. A decentralized Privacy-sensitive Video Surveillance FrameworkDSP, 2013

 

 

Crowd Context-Dependent Privacy Protection Filters

Lupe

While various privacy protection filters have been proposed in the literature, little importance has been given to the context relevance of these filters. In this paper, we specifically focus on the dependency between privacy preservation and crowd density. We show that information about the crowd density in a scene can be used in order to adjust the level of privacy protection according to the local needs. For the estimation of density maps, we use an approach based on FAST feature extraction and local optical flow computation which allow excluding feature points on the background. This process is favorable for the later density function estimation since the influence of features irrelevant to the crowd density is removed. Afterwards, we adapt the protection level of personal privacy in videos according to the crowd density. 

Publication

Fradi, H., Eiselein, V., Keller, I., Dugelay, J.-L., Sikora, T. Crowd Context-Dependent Privacy Protection FiltersDSP, 2013

 

 

Video Indexing and Summarization as a Tool for Privacy Protection

Lupe

The ever increasing number of surveillance camera networks being deployed all over the world has resulted in a high interest in the development of algorithms to automatically analyze the video footage, but has also opened new questions as how to efficiently manage the vast amount of information generated and, more important, how to protect the privacy of the individuals being recorded in their daily life. In this paper, we present a survey on video summarization techniques developed in order to efficiently access to the points of interest in the video footage. Thereby, we emphasize on the links that these techniques show with the task of privacy protection and draw lines of future research directions to incorporate indexing and summarization as tools for privacy protection by design.

Publication

Heras Evangelio, R., Senst, T., Keller, I., Sikora, T. Video Indexing and Summarization as a Tool for Privacy Protection, DSP, 2013

 

 

Navigation

Direktzugang

Schnellnavigation zur Seite über Nummerneingabe