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

Masked point modeling has become a promising scheme of self-supervisedpre-training for point clouds. Existing methods reconstruct either the originalpoints or related features as the objective of pre-training. However,considering the diversity of downstream tasks, it is necessary for the model tohave both low- and high-level representation modeling capabilities to capturegeometric details and semantic contexts during pre-training. To this end,M$^3$CS is proposed to enable the model with the above abilities. Specifically,with masked point cloud as input, M$^3$CS introduces two decoders to predictmasked representations and the original points simultaneously. While an extradecoder doubles parameters for the decoding process and may lead tooverfitting, we propose siamese decoders to keep the amount of learnableparameters unchanged. Further, we propose an online codebook projectingcontinuous tokens into discrete ones before reconstructing masked points. Insuch way, we can enforce the decoder to take effect through the combinations oftokens rather than remembering each token. Comprehensive experiments show thatM$^3$CS achieves superior performance at both classification and segmentationtasks, outperforming existing methods.

M$^3$CS: Multi-Target Masked Point Modeling with Learnable Codebook and Siamese Decoders