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 tackle the data scarcity challenge in few-shot point cloud recognition of3D objects by using a joint prediction from a conventional 3D model and awell-trained 2D model. Surprisingly, such an ensemble, though seems trivial,has hardly been shown effective in recent 2D-3D models. We find out the crux isthe less effective training for the ''joint hard samples'', which have highconfidence prediction on different wrong labels, implying that the 2D and 3Dmodels do not collaborate well. To this end, our proposed invariant trainingstrategy, called InvJoint, does not only emphasize the training more on thehard samples, but also seeks the invariance between the conflicting 2D and 3Dambiguous predictions. InvJoint can learn more collaborative 2D and 3Drepresentations for better ensemble. Extensive experiments on 3D shapeclassification with widely adopted ModelNet10/40, ScanObjectNN and Toys4K, andshape retrieval with ShapeNet-Core validate the superiority of our InvJoint.

Invariant Training 2D-3D Joint Hard Samples for Few-Shot Point Cloud Recognition