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

Weakly supervised point cloud semantic segmentation methods that require 1\%or fewer labels, hoping to realize almost the same performance as fullysupervised approaches, which recently, have attracted extensive researchattention. A typical solution in this framework is to use self-training orpseudo labeling to mine the supervision from the point cloud itself, but ignorethe critical information from images. In fact, cameras widely exist in LiDARscenarios and this complementary information seems to be greatly important for3D applications. In this paper, we propose a novel cross-modality weaklysupervised method for 3D segmentation, incorporating complementary informationfrom unlabeled images. Basically, we design a dual-branch network equipped withan active labeling strategy, to maximize the power of tiny parts of labels anddirectly realize 2D-to-3D knowledge transfer. Afterwards, we establish across-modal self-training framework in an Expectation-Maximum (EM) perspective,which iterates between pseudo labels estimation and parameters updating. In theM-Step, we propose a cross-modal association learning to mine complementarysupervision from images by reinforcing the cycle-consistency between 3D pointsand 2D superpixels. In the E-step, a pseudo label self-rectification mechanismis derived to filter noise labels thus providing more accurate labels for thenetworks to get fully trained. The extensive experimental results demonstratethat our method even outperforms the state-of-the-art fully supervisedcompetitors with less than 1\% actively selected annotations.

Image Understands Point Cloud: Weakly Supervised 3D Semantic Segmentation via Association Learning