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

Person re-identification (re-ID) aims at matching images of the same personacross camera views. Due to varying distances between cameras and persons ofinterest, resolution mismatch can be expected, which would degrade re-IDperformance in real-world scenarios. To overcome this problem, we propose anovel generative adversarial network to address cross-resolution person re-ID,allowing query images with varying resolutions. By advancing adversariallearning techniques, our proposed model learns resolution-invariant imagerepresentations while being able to recover the missing details inlow-resolution input images. The resulting features can be jointly applied forimproving re-ID performance due to preserving resolution invariance andrecovering re-ID oriented discriminative details. Extensive experimentalresults on five standard person re-ID benchmarks confirm the effectiveness ofour method and the superiority over the state-of-the-art approaches, especiallywhen the input resolutions are not seen during training. Furthermore, theexperimental results on two vehicle re-ID benchmarks also confirm thegeneralization of our model on cross-resolution visual tasks. The extensions ofsemi-supervised settings further support the use of our proposed approach toreal-world scenarios and applications.

Cross-Resolution Adversarial Dual Network for Person Re-Identification and Beyond