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 recent years, vision language pre-training frameworks have madesignificant progress in natural language processing and computer vision,achieving remarkable performance improvement on various downstream tasks.However, when extended to point cloud data, existing works mainly focus onbuilding task-specific models, and fail to extract universal 3D vision-languageembedding that generalize well. We carefully investigate three common tasks insemantic 3D scene understanding, and derive key insights into the developmentof a pre-training model. Motivated by these observations, we propose avision-language pre-training framework 3DVLP (3D vision-language pre-trainingwith object contrastive learning), which transfers flexibly on 3Dvision-language downstream tasks. 3DVLP takes visual grounding as the proxytask and introduces Object-level IoU-guided Detection (OID) loss to obtainhigh-quality proposals in the scene. Moreover, we design Object-levelCross-Contrastive alignment (OCC) task and Object-level Self-Contrastivelearning (OSC) task to align the objects with descriptions and distinguishdifferent objects in the scene, respectively. Extensive experiments verify theexcellent performance of 3DVLP on three 3D vision-language tasks, reflectingits superiority in semantic 3D scene understanding.

Vision-Language Pre-training with Object Contrastive Learning for 3D Scene Understanding