The authors propose a general framework for the de-identification of images which subsumes several previously introduced approaches.
A wide range of technological advances have helped to make extensive image and video acquisition close to effortless. As a consequence, many applications which capture image data of people for either immediate inspection or storage and subsequent sharing have become possible. Along with these improved recording capabilities, however, come concerns about the privacy of people visible in the scene. While algorithms have been proposed to de-identify images, currently available methods are still lacking. Unlike the ad-hoc methods currently used in the field the algorithms proposed by the authors aim at providing privacy guarantees. In experiments on illumination-and expression-variant face datasets, the authors show that the proposed algorithms achieve the desired privacy protection while minimally distorting the data. (Publisher abstract provided)
Downloads
Similar Publications
- A Concurrent Tri-Band Low-Noise Amplifier with a Novel Tri-Band Load Resonator Employing Feedback Notches
- Variation Trained Drowsy Cache (VTD-Cache): A History Trained Variation Aware Drowsy Cache for Fine Grain Voltage Scaling
- GeoVisuals: a visual analytics approach to leverage the potential of spatial videos and associated geonarratives