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

Recently, learning-based algorithms have achieved promising performance oncross-spectral image patch matching, which, however, is still far fromsatisfactory for practical application. On the one hand, a lack of large-scaledataset with diverse scenes haunts its further improvement for learning-basedalgorithms, whose performances and generalization rely heavily on the datasetsize and diversity. On the other hand, more emphasis has been put on featurerelation in the spatial domain whereas the scale dependency between featureshas often been ignored, leading to performance degeneration especially whenencountering significant appearance variations for cross-spectral patches. Toaddress these issues, we publish, to be best of our knowledge, the largestvisible and Long-wave Infrared (LWIR) image patch matching dataset, termedVL-CMIM, which contains 1300 pairs of strictly aligned visible and LWIR imagesand over 2 million patch pairs covering diverse scenes such as asteroid, field,country, build, street and water.In addition, a multi-domain feature relationlearning network (MD-FRN) is proposed. Input by the features extracted from afour-branch network, both feature relations in spatial and scale domains arelearned via a spatial correlation module (SCM) and multi-scale adaptiveaggregation module (MSAG), respectively. To further aggregate the multi-domainrelations, a deep domain interactive mechanism (DIM) is applied, where thelearnt spatial-relation and scale-relation features are exchanged and furtherinput into MSCRM and SCM. This mechanism allows our model to learn interactivecross-domain feature relations, leading to improved robustness to significantappearance changes due to different modality.

Learning multi-domain feature relation for visible and Long-wave Infrared image patch matching