paper-with-me

Papers

PointGAC: Geometric-Aware Codebook for Masked Point Cloud Modeling

2025-07-07 · Abiao Li, Chenlei Lv, Yuming Fang, Yifan Zuo, Jian Zhang, Guofeng Mei arxiv

Most masked point cloud modeling (MPM) methods follow a regression paradigm to reconstruct the coordinate or feature of masked regions. However, they tend to over-constrain the model to learn the details of the masked region, resulting in failure to capture generalized features. To address this limitation, we propose \textbf{\textit{PointGAC}}, a novel clustering-based MPM method that aims to align the feature distribution of masked regions. Specially, it features an online codebook-guided teacher-student framework. Firstly, it presents a geometry-aware partitioning strategy to extract initial patches. Then, the teacher model updates a codebook via online k-means based on features extracted from the complete patches. This procedure facilitates codebook vectors to become cluster centers. Afterward, we assigns the unmasked features to their corresponding cluster centers, and the student model aligns the assignment for the reconstructed masked features. This strategy focuses on identifying the cluster centers to which the masked features belong, enabling the model to learn more generalized feature representations. Benefiting from a proposed codebook maintenance mechanism, codebook vectors are actively updated, which further increases the efficiency of semantic feature learning. Experiments validate the effectiveness of the proposed method on various downstream tasks. Code is available at https://github.com/LAB123-tech/PointGAC

📄 PDF Abstract BibTeX arXiv:2507.04801

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

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

2023-09-23 · Qibo Qiu, Honghui Yang, Wenxiao Wang, Shun Zhang 외

Masked point modeling has become a promising scheme of self-supervised pre-training for point clouds. Existing methods reconstruct either the original points or related features as the objective of pre-training. However,…

DecoderDiversity

Efficient and Explainable End-to-End Autonomous Driving via Masked Vision-Language-Action Diffusion

2026-02-24 · Jiaru Zhang, Manav Gagvani, Can Cui, Juntong Peng 외 arxiv

Large Language Models (LLMs) and Vision-Language Models (VLMs) have emerged as promising candidates for end-to-end autonomous driving. However, these models typically face challenges in inference latency, action precisio…

Autonomous Driving

MAPE-PPI: Towards Effective and Efficient Protein-Protein Interaction Prediction via Microenvironment-Aware Protein Embedding

2024-02-22 · Lirong Wu, Yijun Tian, Yufei Huang, Siyuan Li 외

Protein-Protein Interactions (PPIs) are fundamental in various biological processes and play a key role in life activities. The growing demand and cost of experimental PPI assays require computational methods for efficie…

Computational EfficiencyPrediction

CodedVTR: Codebook-based Sparse Voxel Transformer with Geometric Guidance

2022-03-18 · CVPR 2022 1 · Tianchen Zhao, Niansong Zhang, Xuefei Ning, He Wang 외

Transformers have gained much attention by outperforming convolutional neural networks in many 2D vision tasks. However, they are known to have generalization problems and rely on massive-scale pre-training and sophistic…

3D Semantic SegmentationSemantic Segmentation

Rethinking End-to-End 2D to 3D Scene Segmentation in Gaussian Splatting

2025-03-18 · CVPR 2025 1 · Runsong Zhu, Shi Qiu, Zhengzhe Liu, Ka-Hei Hui 외

Lifting multi-view 2D instance segmentation to a radiance field has proven to be effective to enhance 3D understanding. Existing methods rely on direct matching for end-to-end lifting, yielding inferior results; or emplo…

Instance SegmentationObjectScene SegmentationSegmentation+1