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

Dramatic appearance variation due to pose constitutes a great challenge infine-grained recognition, one which recent methods using attention mechanismsor second-order statistics fail to adequately address. Modern CNNs typicallylack an explicit understanding of object pose and are instead confused byentangled pose and appearance. In this paper, we propose a unified objectrepresentation built from a hierarchy of pose-aligned regions. Rather thanrepresenting an object by regions aligned to image axes, the proposedrepresentation characterizes appearance relative to the object's pose usingpose-aligned patches whose features are robust to variations in pose, scale androtation. We propose an algorithm that performs pose estimation and forms theunified object representation as the concatenation of hierarchical pose-alignedregions features, which is then fed into a classification network. The proposedalgorithm surpasses the performance of other approaches, increasing thestate-of-the-art by nearly 2% on the widely-used CUB-200 dataset and by morethan 8% on the much larger NABirds dataset. The effectiveness of this paradigmrelative to competing methods suggests the critical importance of disentanglingpose and appearance for continued progress in fine-grained recognition.

Aligned to the Object, not to the Image: A Unified Pose-aligned Representation for Fine-grained Recognition