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

Generalizable 3D part segmentation is important but challenging in vision androbotics. Training deep models via conventional supervised methods requireslarge-scale 3D datasets with fine-grained part annotations, which are costly tocollect. This paper explores an alternative way for low-shot part segmentationof 3D point clouds by leveraging a pretrained image-language model, GLIP, whichachieves superior performance on open-vocabulary 2D detection. We transfer therich knowledge from 2D to 3D through GLIP-based part detection on point cloudrendering and a novel 2D-to-3D label lifting algorithm. We also utilizemulti-view 3D priors and few-shot prompt tuning to boost performancesignificantly. Extensive evaluation on PartNet and PartNet-Mobility datasetsshows that our method enables excellent zero-shot 3D part segmentation. Ourfew-shot version not only outperforms existing few-shot approaches by a largemargin but also achieves highly competitive results compared to the fullysupervised counterpart. Furthermore, we demonstrate that our method can bedirectly applied to iPhone-scanned point clouds without significant domaingaps.

PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Models