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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Do object part localization methods produce bilaterally symmetric results onmirror images? Surprisingly not, even though state of the art methods augmentthe training set with mirrored images. In this paper we take a closer look intothis issue. We first introduce the concept of mirrorability as the ability of amodel to produce symmetric results in mirrored images and introduce acorresponding measure, namely the \textit{mirror error} that is defined as thedifference between the detection result on an image and the mirror of thedetection result on its mirror image. We evaluate the mirrorability of severalstate of the art algorithms in two of the most intensively studied problems,namely human pose estimation and face alignment. Our experiments lead toseveral interesting findings: 1) Surprisingly, most of state of the art methodsstruggle to preserve the mirror symmetry, despite the fact that they do havevery similar overall performance on the original and mirror images; 2) the lowmirrorability is not caused by training or testing sample bias - all algorithmsare trained on both the original images and their mirrored versions; 3) themirror error is strongly correlated to the localization/alignment error (withcorrelation coefficients around 0.7). Since the mirror error is calculatedwithout knowledge of the ground truth, we show two interesting applications -in the first it is used to guide the selection of difficult samples and in thesecond to give feedback in a popular Cascaded Pose Regression method for facealignment.

Mirror, mirror on the wall, tell me, is the error small?