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
Hanxiao Tan
Summary: So far, few researchers have studied the explainability of point cloud neural networks. Traditional activation maximization methods fail to generate human-perceptible global explanations for these networks. We propose new generative model-based activation maximization approaches to clearly outline the global explanations and improve their comprehensibility. Our experiments demonstrate that our approaches outperform regularization-based ones both qualitatively and quantitatively. This is the first work investigating the global explainability of point cloud networks.
2023 IEEE/CVF WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV)
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhaoyang Xia et al.
Summary: Training deep models for semantic scene completion is challenging due to sparse and incomplete input, diverse objects of different scales, and label noise for moving objects. To address these challenges, the proposed method includes a redesign of the completion sub-network, distilling knowledge from a multi-frame model, and completion label rectification. The experiments conducted on two benchmarks demonstrate the effectiveness and superiority of the method.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhikai Chen et al.
Summary: This paper presents a novel shape completion architecture called AnchorFormer, which leverages pattern-aware discriminative nodes (anchors) to capture regional information of objects and uses a modulation scheme to transform sparse points into detailed 3D structures. Experimental results demonstrate the superiority of AnchorFormer over existing techniques in point cloud completion.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Sangmin Hong et al.
Summary: This paper proposes a self-supervised framework ACL-SPC for point cloud completion, which can be trained and tested on the same data. Extensive experiments demonstrate the effectiveness of ACL-SPC and justify the necessity of self-supervised learning in the point cloud completion task.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Shanshan Li et al.
Summary: This paper proposes a novel point cloud completion approach called ProxyFormer, which utilizes proxies to communicate information and recover complete point clouds from partial ones. Experimental results demonstrate that our method outperforms state-of-the-art completion networks on several benchmark datasets and has the fastest inference speed.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Changfeng Ma et al.
Summary: In this paper, we propose an Unsupervised Symmetric Shape-Preserving Autoencoding Network (USSPA) to predict complete point clouds of objects from real scenes. Our method preserves input shapes, predicts accurate results, and adapts to multi-category data through symmetry learning and carefully designed upsampling refinement.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Junming Zhang et al.
Summary: This paper proposes a hyperspherical module for point cloud completion task, which transforms and normalizes embeddings to improve generalization results during testing and enable more stable training with compact embedding distributions.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR
(2023)
Article
Environmental Sciences
Shikun Li et al.
Summary: With the development of scanning technology, point cloud registration plays a significant role in various fields. Traditional algorithms like ICP search for corresponding points based on the closest point, while recent deep learning-based algorithms utilize deep features for point calculation. However, the partiality of point clouds poses challenges for partial-to-partial registration. To address this, we propose VPRNet, which combines a virtual point generation network with a registration network to overcome the limitations of partiality. The experiments demonstrate the superior performance of our proposed algorithm compared to existing methods.
Proceedings Paper
Computer Science, Artificial Intelligence
Junshu Tang et al.
Summary: This paper introduces a novel topology-aware point cloud completion model LAKe-Net, which aims to tackle missing topology by localizing aligned keypoints and predicting keypoints-skeleton-shape. Experimental results demonstrate that this method achieves state-of-the-art performance in point cloud completion.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Alexandre Boulch et al.
Summary: Implicit neural networks have been successfully used for surface reconstruction from point clouds, but many of them face scalability issues. To overcome this limitation, we propose a method that uses point cloud convolutions and learning-based interpolation to reconstruct the surface of objects better, producing finer details.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Yingjie Cai et al.
Summary: The study proposes a novel framework to enhance the completeness of point cloud by learning a unified and structured latent space. A series of constraints are adopted to promote the learning of such a structured space, with experiments showing that the method outperforms state-of-the-art unsupervised methods on multiple datasets.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Article
Environmental Sciences
Ming Wei et al.
Summary: The application of 3D scenes has been expanding in recent years, but the reliability of 3D point clouds acquired using sensors is limited, causing difficulties in their utilization. To address this issue, point cloud completion techniques can reconstruct and restore sparse and incomplete point clouds to enhance their realism. In this study, we propose a cyclic global guiding network structure that considers both local details and overall characteristics of the whole cloud for point cloud completion tasks. We introduce fitting planes and layered folding attention modules based on global guidance to strengthen the local effect. Experimental results demonstrate the effectiveness of our method on diverse datasets and its superiority over other networks.
Proceedings Paper
Computer Science, Artificial Intelligence
Liang Pan et al.
Summary: The paper introduces a variational framework, Variational Relational point Completion network (VRCNet), with probabilistic modeling and relational enhancement properties to address the challenges of incomplete point clouds and lack of local details. The method outperforms state-of-the-art approaches in standard point cloud completion benchmarks and demonstrates great generalizability and robustness in real-world point cloud scans.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Xin Wen et al.
Summary: This paper introduces a new perspective on point cloud completion task by formulating the prediction as a point cloud deformation process using a novel neural network named PMP-Net. PMP-Net predicts unique point moving paths for each point to improve the quality of the predicted complete shape through shortest total point moving distances. Experimental results on Completion3D and PCN datasets show advantages over state-of-the-art point cloud completion methods.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Article
Robotics
Liang Pan
IEEE ROBOTICS AND AUTOMATION LETTERS
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Xin Wen et al.
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2020)
Article
Computer Science, Software Engineering
Yue Wang et al.
ACM TRANSACTIONS ON GRAPHICS
(2019)
Article
Computer Science, Software Engineering
Manuele Sabbadin et al.
COMPUTER GRAPHICS FORUM
(2019)
Proceedings Paper
Automation & Control Systems
Haohao Huang et al.
PROCEEDINGS OF THE 2019 31ST CHINESE CONTROL AND DECISION CONFERENCE (CCDC 2019)
(2019)
Proceedings Paper
Engineering, Electrical & Electronic
Shubham Agrawal et al.
2019 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Lyne P. Tchapmi et al.
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019)
(2019)
Article
Engineering, Electrical & Electronic
Xiaoshui Huang et al.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2018)
