相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Proceedings Paper
Computer Science, Artificial Intelligence
Minmin Yang et al.
Summary: Recent years have seen significant progress in few-shot image classification, but few-shot 3D point cloud classification is still limited. Real-world 3D point cloud data is affected by occlusions, noise, and deformation, making few-shot 3D point cloud classification even more challenging. This paper proposes a cross-modality feature fusion network for few-shot 3D point cloud classification, which achieves better performance even with point cloud data missing points. Two parallel models, a projection-based model with ResNet-18 and a point-based model with DGCNN backbone, are trained. The Support-Query Mutual Attention (sqMA) module is designed to exploit the correlation between support and query features. Extensive experiments on three datasets demonstrate the effectiveness and robustness of the proposed method. It outperforms state-of-the-art baselines, especially on the challenging ScanObjectNN dataset with missing points.
2023 IEEE/CVF WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV)
(2023)
Article
Environmental Sciences
Weipeng Jing et al.
Summary: Classification and segmentation of point clouds face challenges in identifying attentional regions, leading to the proposal of AGNet which extracts spatial information efficiently and selects important features accurately. The model shows advantages in point set classification and segmentation tasks according to both quantitative and qualitative experiments.
Article
Computer Science, Artificial Intelligence
Shaohua Qi et al.
Summary: This study proposes a double-weighting convolutional neural network method based on the L2-S grouping mechanism to improve the accuracy of 3D object recognition based on multiple views. By designing bidirectional long short-term memory module and reasonable grouping mechanism, the extracted features are more representative and achieve state-of-the-art performance.
IET COMPUTER VISION
(2022)
Article
Environmental Sciences
Wenju Wang et al.
Summary: This paper proposes a neural network model FSDCNet based on the fusion of static and dynamic convolution for multiview 3D point cloud classification. By designing view selection methods, local feature extraction operators, and lightweight and adaptive dynamic convolution operators, the model improves adaptability and classification accuracy.
Article
Engineering, Civil
Dening Lu et al.
Summary: This paper proposes a hierarchical framework, called 3D Convolution-Transformer Network (3DCTN), which combines convolutions with Transformers for point cloud classification. Our method achieves feature learning on point clouds through the use of local feature aggregating blocks and global feature learning blocks, resulting in state-of-the-art classification performance.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Review
Chemistry, Analytical
Elizar Elizar et al.
Summary: Most existing convolutional neural network-based deep-learning models suffer from spatial information loss and inadequate feature representation due to their inability to capture multiscale-context information and semantic information throughout pooling operations. Multiscale feature learning and fusion are crucial for optimal feature extraction and representation in deep-learning networks.
Article
Computer Science, Artificial Intelligence
Qi Liang et al.
Summary: This study proposes a Multi-view based Hierarchical Fusion Pooling Method (MHFP) for 3D Model Recognition, which hierarchically fuses features of multi-view considering the correlation between views. The approach effectively removes redundant information and retains essential information, with a 3D attention module designed to extract correlation between views. Experimental results demonstrate the superiority of the proposed method in 3D shape recognition.
PATTERN RECOGNITION LETTERS
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Jianwen Xie et al.
Summary: The study introduces a generative model for unordered point sets, capable of learning coordinate encodings for individual points and aggregating features for the entire point cloud, enabling point cloud generation and reconstruction.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Article
Computer Science, Artificial Intelligence
Tan Yu et al.
Summary: This paper investigates 3D object representation learning from the perspective of set-to-set matching, proposing effective models for object recognition through local features pooling and two pooling methods. The experiments on public benchmark datasets demonstrate the efficacy of the proposed models in 3D object recognition.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Yong Xu et al.
Summary: In recent years, multi-view learning has become a promising approach for 3D shape recognition by identifying shapes based on 2D views from different angles. This paper proposes a correspondence-aware representation (CAR) module that finds potential intra-view and cross-view correspondences through kNN search in semantic space and aggregates shape features via learned transforms. Incorporating the CAR module into a ResNet-18 backbone, an effective deep model called CAR-Net is introduced for 3D shape classification and retrieval, demonstrating the effectiveness and excellent performance of the CAR module.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Zhizhong Han et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2019)
Article
Computer Science, Artificial Intelligence
Zhizhong Han et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2019)
Article
Robotics
Yizhak Ben-Shabat et al.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
Tan Yu et al.
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
Pingping Zhang et al.
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Qibin Hou et al.
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017)
(2017)
Article
Computer Science, Software Engineering
Nate Hagbi et al.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2011)