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

This work focuses on mitigating two limitations in the joint learning oflocal feature detectors and descriptors. First, the ability to estimate thelocal shape (scale, orientation, etc.) of feature points is often neglectedduring dense feature extraction, while the shape-awareness is crucial toacquire stronger geometric invariance. Second, the localization accuracy ofdetected keypoints is not sufficient to reliably recover camera geometry, whichhas become the bottleneck in tasks such as 3D reconstruction. In this paper, wepresent ASLFeat, with three light-weight yet effective modifications tomitigate above issues. First, we resort to deformable convolutional networks todensely estimate and apply local transformation. Second, we take advantage ofthe inherent feature hierarchy to restore spatial resolution and low-leveldetails for accurate keypoint localization. Finally, we use a peakinessmeasurement to relate feature responses and derive more indicative detectionscores. The effect of each modification is thoroughly studied, and theevaluation is extensively conducted across a variety of practical scenarios.State-of-the-art results are reported that demonstrate the superiority of ourmethods.

ASLFeat: Learning Local Features of Accurate Shape and Localization