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

Animal pose estimation has recently come into the limelight due to itsapplication in biology, zoology, and aquaculture. Deep learning methods haveeffectively been applied to human pose estimation. However, the majorbottleneck to the application of these methods to animal pose estimation is theunavailability of sufficient quantities of labeled data. Though there are amplequantities of unlabelled data publicly available, it is economicallyimpractical to label large quantities of data for each animal. In addition, dueto the wide variety of body shapes in the animal kingdom, the transfer ofknowledge across domains is ineffective. Given the fact that the human brain isable to recognize animal pose without requiring large amounts of labeled data,it is only reasonable that we exploit unsupervised learning to tackle theproblem of animal pose recognition from the available, unlabelled data. In thispaper, we introduce a novel architecture that is able to recognize the pose ofmultiple animals fromunlabelled data. We do this by (1) removing backgroundinformation from each image and employing an edge detection algorithm on thebody of the animal, (2) Tracking motion of the edge pixels and performingagglomerative clustering to segment body parts, (3) employing contrastivelearning to discourage grouping of distant body parts together. Hence we areable to distinguish between body parts of the animal, based on their visualbehavior, instead of the underlying anatomy. Thus, we are able to achieve amore effective classification of the data than their human-labeledcounterparts. We test our model on the TigDog and WLD (WildLife Documentary)datasets, where we outperform state-of-the-art approaches by a significantmargin. We also study the performance of our model on other public data todemonstrate the generalization ability of our model.

Pose Recognition in the Wild: Animal pose estimation using Agglomerative Clustering and Contrastive Learning