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

The classical matching pipeline used for visual localization typicallyinvolves three steps: (i) local feature detection and description, (ii) featurematching, and (iii) outlier rejection. Recently emerged correspondence networkspropose to perform those steps inside a single network but suffer from lowmatching resolution due to the memory bottleneck. In this work, we propose anew perspective to estimate correspondences in a detect-to-refine manner, wherewe first predict patch-level match proposals and then refine them. We presentPatch2Pix, a novel refinement network that refines match proposals byregressing pixel-level matches from the local regions defined by thoseproposals and jointly rejecting outlier matches with confidence scores.Patch2Pix is weakly supervised to learn correspondences that are consistentwith the epipolar geometry of an input image pair. We show that our refinementnetwork significantly improves the performance of correspondence networks onimage matching, homography estimation, and localization tasks. In addition, weshow that our learned refinement generalizes to fully-supervised methodswithout re-training, which leads us to state-of-the-art localizationperformance. The code is available at https://github.com/GrumpyZhou/patch2pix.

Patch2Pix: Epipolar-Guided Pixel-Level Correspondences