English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Over the course of the last decade, infrared (IR) and particularly thermal IRimaging based face recognition has emerged as a promising complement toconventional, visible spectrum based approaches which continue to struggle whenapplied in the real world. While inherently insensitive to visible spectrumillumination changes, IR images introduce specific challenges of their own,most notably sensitivity to factors which affect facial heat emission patterns,e.g. emotional state, ambient temperature, and alcohol intake. In addition,facial expression and pose changes are more difficult to correct in IR imagesbecause they are less rich in high frequency detail which is an important cuefor fitting any deformable model. We describe a novel method which addressesthese challenges. To normalize for pose and facial expression changes wegenerate a synthetic frontal image of a face in a canonical, neutral facialexpression from an image of the face in an arbitrary pose and facialexpression. This is achieved by piecewise affine warping which follows activeappearance model (AAM) fitting. This is the first publication which exploresthe use of an AAM on thermal IR images; we propose a pre-processing step whichenhances detail in thermal images, making AAM convergence faster and moreaccurate. To overcome the problem of thermal IR image sensitivity to thepattern of facial temperature emissions we describe a representation based onreliable anatomical features. In contrast to previous approaches, ourrepresentation is not binary; rather, our method accounts for the reliabilityof the extracted features. This makes the proposed representation much morerobust both to pose and scale changes. The effectiveness of the proposedapproach is demonstrated on the largest public database of thermal IR images offaces on which it achieved 100% identification, significantly outperformingprevious methods.

Vesselness features and the inverse compositional AAM for robust face recognition using thermal IR