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

Visual identification of individual animals that bear unique natural bodymarkings is an important task in wildlife conservation. The photo databases ofanimal markings grow larger and each new observation has to be matched againstthousands of images. Existing photo-identification solutions have constraintson image quality and appearance of the pattern of interest in the image. Theseconstraints limit the use of photos from citizen scientists. We present a novelsystem for visual re-identification based on unique natural markings that isrobust to occlusions, viewpoint and illumination changes. We adapt methodsdeveloped for face re-identification and implement a deep convolutional neuralnetwork (CNN) to learn embeddings for images of natural markings. The distancebetween the learned embedding points provides a dissimilarity measure betweenthe corresponding input images. The network is optimized using the triplet lossfunction and the online semi-hard triplet mining strategy. The proposedre-identification method is generic and not species specific. We evaluate theproposed system on image databases of manta ray belly patterns and humpbackwhale flukes. To be of practical value and adopted by marine biologists, are-identification system needs to have a top-10 accuracy of at least 95%. Theproposed system achieves this performance standard.

Robust Re-identification of Manta Rays from Natural Markings by Learning Pose Invariant Embeddings