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

We introduce PartSTAD, a method designed for the task adaptation of 2D-to-3Dsegmentation lifting. Recent studies have highlighted the advantages ofutilizing 2D segmentation models to achieve high-quality 3D segmentationthrough few-shot adaptation. However, previous approaches have focused onadapting 2D segmentation models for domain shift to rendered images andsynthetic text descriptions, rather than optimizing the model specifically for3D segmentation. Our proposed task adaptation method finetunes a 2D boundingbox prediction model with an objective function for 3D segmentation. Weintroduce weights for 2D bounding boxes for adaptive merging and learn theweights using a small additional neural network. Additionally, we incorporateSAM, a foreground segmentation model on a bounding box, to improve theboundaries of 2D segments and consequently those of 3D segmentation. Ourexperiments on the PartNet-Mobility dataset show significant improvements withour task adaptation approach, achieving a 7.0%p increase in mIoU and a 5.2%pimprovement in mAP_50 for semantic and instance segmentation compared to theSotA few-shot 3D segmentation model.

PartSTAD: 2D-to-3D Part Segmentation Task Adaptation