The BERT-style (Bidirectional Encoder Representations from Transformers) pre-training paradigm has achieved remarkable success in both NLP (Natural Language Processing) and CV (Computer Vision). However, due to the sparsity and lack of semantic information in point cloud data, directly applying this paradigm faces significant obstacles. To address these issues, we propose LSV-MAE, a masked autoencoding pre-training scheme designed for voxel representations. We pre-train the backbone to reconstruct the masked voxels features extracted by PointNN. To enhance the feature extraction capability of the encoder, the point cloud is voxelized with different voxel sizes at different pre-training stages. Meanwhile, to minimize the effect of masking key points during the masking stage, the masked voxel features is re-integrated during the decoder processing. To test the proposed approach, experiments are conducted on well-known datasets. It is shown that the proposed method can improve detection accuracy by 1% to 18%, compared to the model without pre-training, across datasets of different sizes.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Rethinking Masked-Autoencoder-Based 3D Point Cloud Pretraining


    Contributors:
    Cheng, Nuo (author) / Luo, Chuanyu (author) / Li, Xinzhe (author) / Hu, Ruizhi (author) / Li, Han (author) / Ma, Sikun (author) / Ren, Zhong (author) / Jiang, Haipeng (author) / Li, Xiaohan (author) / Lei, Shengguang (author)


    Publication date :

    2024-06-02


    Size :

    1338115 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    APS: ACTIVE PRETRAINING WITH SUCCESSOR FEATURES

    LIU, HAO / ABBEEL, PIETER | TIBKAT | 2022


    The point - Rethinking luxury

    Cumberford, Robert | Online Contents | 2001


    Rethinking the White Point

    Farrell, J. E. / Wandell, B. A. / Society for Information Display et al. | British Library Conference Proceedings | 1993


    Score-Based Multimodal Autoencoder

    Wesego, Daniel / Rooshenas, Pedram | ArXiv | 2023

    Free access