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

In 3D Referring Expression Segmentation (3D-RES), the earlier approach adoptsa two-stage paradigm, extracting segmentation proposals and then matching themwith referring expressions. However, this conventional paradigm encounterssignificant challenges, most notably in terms of the generation of lacklusterinitial proposals and a pronounced deceleration in inference speed. Recognizingthese limitations, we introduce an innovative end-to-end Superpoint-TextMatching Network (3D-STMN) that is enriched by dependency-driven insights. Oneof the keystones of our model is the Superpoint-Text Matching (STM) mechanism.Unlike traditional methods that navigate through instance proposals, STMdirectly correlates linguistic indications with their respective superpoints,clusters of semantically related points. This architectural decision empowersour model to efficiently harness cross-modal semantic relationships, primarilyleveraging densely annotated superpoint-text pairs, as opposed to the moresparse instance-text pairs. In pursuit of enhancing the role of text in guidingthe segmentation process, we further incorporate the Dependency-DrivenInteraction (DDI) module to deepen the network's semantic comprehension ofreferring expressions. Using the dependency trees as a beacon, this modulediscerns the intricate relationships between primary terms and their associateddescriptors in expressions, thereby elevating both the localization andsegmentation capacities of our model. Comprehensive experiments on theScanRefer benchmark reveal that our model not only set new performancestandards, registering an mIoU gain of 11.7 points but also achieve astaggering enhancement in inference speed, surpassing traditional methods by95.7 times. The code and models are available athttps://github.com/sosppxo/3D-STMN.

3D-STMN: Dependency-Driven Superpoint-Text Matching Network for End-to-End 3D Referring Expression Segmentation