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

Line segments are ubiquitous in our human-made world and are increasinglyused in vision tasks. They are complementary to feature points thanks to theirspatial extent and the structural information they provide. Traditional linedetectors based on the image gradient are extremely fast and accurate, but lackrobustness in noisy images and challenging conditions. Their learnedcounterparts are more repeatable and can handle challenging images, but at thecost of a lower accuracy and a bias towards wireframe lines. We propose tocombine traditional and learned approaches to get the best of both worlds: anaccurate and robust line detector that can be trained in the wild withoutground truth lines. Our new line segment detector, DeepLSD, processes imageswith a deep network to generate a line attraction field, before converting itto a surrogate image gradient magnitude and angle, which is then fed to anyexisting handcrafted line detector. Additionally, we propose a new optimizationtool to refine line segments based on the attraction field and vanishingpoints. This refinement improves the accuracy of current deep detectors by alarge margin. We demonstrate the performance of our method on low-level linedetection metrics, as well as on several downstream tasks using multiplechallenging datasets. The source code and models are available athttps://github.com/cvg/DeepLSD.

DeepLSD: Line Segment Detection and Refinement with Deep Image Gradients