scholarly journals AttPNet: Attention-Based Deep Neural Network for 3D Point Set Analysis

Sensors ◽  
2020 ◽  
Vol 20 (19) ◽  
pp. 5455
Author(s):  
Yufeng Yang ◽  
Yixiao Ma ◽  
Jing Zhang ◽  
Xin Gao ◽  
Min Xu
Keyword(s):  
Point Cloud ◽  
3D Structure ◽  
Data Availability ◽  
Major Type ◽  
Global Features ◽  
Structure Representation ◽  
Fine Grained ◽  
Point Set ◽  
Classification Tasks ◽  
Novel Model

Point set is a major type of 3D structure representation format characterized by its data availability and compactness. Most former deep learning-based point set models pay equal attention to different point set regions and channels, thus having limited ability in focusing on small regions and specific channels that are important for characterizing the object of interest. In this paper, we introduce a novel model named Attention-based Point Network (AttPNet). It uses attention mechanism for both global feature masking and channel weighting to focus on characteristic regions and channels. There are two branches in our model. The first branch calculates an attention mask for every point. The second branch uses convolution layers to abstract global features from point sets, where channel attention block is adapted to focus on important channels. Evaluations on the ModelNet40 benchmark dataset show that our model outperforms the existing best model in classification tasks by 0.7% without voting. In addition, experiments on augmented data demonstrate that our model is robust to rotational perturbations and missing points. We also design a Electron Cryo-Tomography (ECT) point cloud dataset and further demonstrate our model’s ability in dealing with fine-grained structures on the ECT dataset.

scholarly journals Multiscale Receptive Fields Graph Attention Network for Point Cloud Classification

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Xi-An Li ◽  
Li-Yan Wang ◽  
Jian Lu
Keyword(s):  
Point Cloud ◽  
Receptive Fields ◽  
Semantic Features ◽  
Attention Network ◽  
Fine Grained ◽  
Local Patch ◽  
Classification Tasks ◽  
Point Cloud Classification ◽  
Cloud Process ◽  
Breaking Work

Understanding the implication of point cloud is still challenging in the aim of classification or segmentation for point cloud due to its irregular and sparse structure. As we have known, PointNet architecture as a ground-breaking work for point cloud process can learn shape features directly on unordered 3D point cloud and has achieved favorable performance, such as 86% mean accuracy and 89.2% overall accuracy for classification task, respectively. However, this model fails to consider the fine-grained semantic information of local structure for point cloud. Then, a multiscale receptive fields graph attention network (named after MRFGAT) by means of semantic features of local patch for point cloud is proposed in this paper, and the learned feature map for our network can well capture the abundant features information of point cloud. The proposed MRFGAT architecture is tested on ModelNet datasets, and results show it achieves state-of-the-art performance in shape classification tasks, such as it outperforms GAPNet (Chen et al.) model by 0.1% in terms of OA and compete with DGCNN (Wang et al.) model in terms of MA.

scholarly journals Explicit Interaction Model towards Text Classification

2019 ◽  
Vol 33 ◽  
pp. 6359-6366 ◽  
Author(s):  
Cunxiao Du ◽  
Zhaozheng Chen ◽  
Fuli Feng ◽  
Lei Zhu ◽  
Tian Gan ◽  
...  
Keyword(s):  
Language Processing ◽  
Text Classification ◽  
Deep Neural Networks ◽  
Interaction Mechanism ◽  
Interaction Model ◽  
Classification Task ◽  
Fine Grained ◽  
Word Level ◽  
Benchmark Datasets ◽  
Classification Tasks

Text classification is one of the fundamental tasks in natural language processing. Recently, deep neural networks have achieved promising performance in the text classification task compared to shallow models. Despite of the significance of deep models, they ignore the fine-grained (matching signals between words and classes) classification clues since their classifications mainly rely on the text-level representations. To address this problem, we introduce the interaction mechanism to incorporate word-level matching signals into the text classification task. In particular, we design a novel framework, EXplicit interAction Model (dubbed as EXAM), equipped with the interaction mechanism. We justified the proposed approach on several benchmark datasets including both multilabel and multi-class text classification tasks. Extensive experimental results demonstrate the superiority of the proposed method. As a byproduct, we have released the codes and parameter settings to facilitate other researches.

Adaptive Slicing of Moving Least Squares Surfaces: Toward Direct Manufacturing of Point Set Surfaces

2007 ◽  
Author(s):  
Pinghai Yang ◽  
Xiaoping Qian
Keyword(s):  
Least Squares ◽  
Point Cloud ◽  
Moving Least Squares ◽  
Point Sets ◽  
New Approach ◽  
Adaptive Slicing ◽  
3D Sensing ◽  
Point Set ◽  
Point Set Surfaces ◽  
Model Conversion

Rapid advancement of 3D sensing techniques has lead to dense and accurate point cloud of an object to be readily available. The growing use of such scanned point sets in product design, analysis and manufacturing necessitates research on direct processing of point set surfaces. In this paper, we present an approach that enables the direct layered manufacturing of point set surfaces. This new approach is based on adaptive slicing of moving least squares (MLS) surfaces. Salient features of this new approach include: 1) it bypasses the laborious surface reconstruction and avoids model conversion induced accuracy loss; 2) the resulting layer thickness and layer contours are adaptive to local curvature and thus it leads to better surface quality and more efficient fabrication; 3) the MLS surface naturally smoothes the point cloud and allows up-sampling and down-sampling, and thus it is robust even for noisy or sparse point sets. Experimental results of the slicing algorithm on both synthetic and scanned point sets are presented.

scholarly journals Self-Sampling for Neural Point Cloud Consolidation

2021 ◽  
Vol 40 (5) ◽  
pp. 1-14
Author(s):  
Gal Metzer ◽  
Rana Hanocka ◽  
Raja Giryes ◽  
Daniel Cohen-Or
Keyword(s):  
Point Cloud ◽  
Sampling Methods ◽  
Difficult Problem ◽  
Local Statistics ◽  
Novel Technique ◽  
Input Point ◽  
Shape Learning ◽  
Fixed Set ◽  
Point Set ◽  
Non Local

We introduce a novel technique for neural point cloud consolidation which learns from only the input point cloud. Unlike other point up-sampling methods which analyze shapes via local patches, in this work, we learn from global subsets. We repeatedly self-sample the input point cloud with global subsets that are used to train a deep neural network. Specifically, we define source and target subsets according to the desired consolidation criteria (e.g., generating sharp points or points in sparse regions). The network learns a mapping from source to target subsets, and implicitly learns to consolidate the point cloud. During inference, the network is fed with random subsets of points from the input, which it displaces to synthesize a consolidated point set. We leverage the inductive bias of neural networks to eliminate noise and outliers, a notoriously difficult problem in point cloud consolidation. The shared weights of the network are optimized over the entire shape, learning non-local statistics and exploiting the recurrence of local-scale geometries. Specifically, the network encodes the distribution of the underlying shape surface within a fixed set of local kernels, which results in the best explanation of the underlying shape surface. We demonstrate the ability to consolidate point sets from a variety of shapes, while eliminating outliers and noise.

scholarly journals SAN-GAL: Spatial Attention Network Guided by Attribute Label for Person Re-identification

2021 ◽  
Vol 2021 ◽  
pp. 1-8
Author(s):  
Shaoqi Hou ◽  
Chunhui Liu ◽  
Kangning Yin ◽  
Yiyin Ding ◽  
Zhiguo Wang ◽  
...  
Keyword(s):  
Spatial Attention ◽  
Binary Classification ◽  
Test Results ◽  
Global Features ◽  
Matching Problem ◽  
Heat Map ◽  
Attention Network ◽  
Fine Grained ◽  
Attribute Classification ◽  
High Degree

Person Re-identification (Re-ID) is aimed at solving the matching problem of the same pedestrian at a different time and in different places. Due to the cross-device condition, the appearance of different pedestrians may have a high degree of similarity; at this time, using the global features of pedestrians to match often cannot achieve good results. In order to solve these problems, we designed a Spatial Attention Network Guided by Attribute Label (SAN-GAL), which is a dual-trace network containing both attribute classification and Re-ID. Different from the previous approach of simply adding a branch of attribute binary classification network, our SAN-GAL is mainly divided into two connecting steps. First, with attribute labels as guidance, we generate Attribute Attention Heat map (AAH) through Grad-CAM algorithm to accurately locate fine-grained attribute areas of pedestrians. Then, the Attribute Spatial Attention Module (ASAM) is constructed according to the AHH which is taken as the prior knowledge and introduced into the Re-ID network to assist in the discrimination of the Re-ID task. In particular, our SAN-GAL network can integrate the local attribute information and global ID information of pedestrians without introducing additional attribute region annotation, which has good flexibility and adaptability. The test results on Market1501 and DukeMTMC-reID show that our SAN-GAL can achieve good results and can achieve 85.8% Rank-1 accuracy on DukeMTMC-reID dataset, which is obviously competitive compared with most Re-ID algorithms.

scholarly journals Ground Segmentation Algorithm of Lidar Point Cloud Based on Ray-Ransac

2021 ◽  
Vol 15 ◽  
pp. 970-977
Author(s):  
Yawei Zhao ◽  
Yanju Liu ◽  
Yang Yu ◽  
Jiawei Zhou
Keyword(s):  
Point Cloud ◽  
Structural Characteristics ◽  
Segmentation Algorithm ◽  
Model Parameters ◽  
Good Precision ◽  
Seed Point ◽  
Original Seed ◽  
Ground Segmentation ◽  
Point Set ◽  
Ransac Algorithm

Aiming at the problems of poor segmentation effect, low efficiency and poor robustness of the Ransac ground segmentation algorithm, this paper proposes a radar segmentation algorithm based on Ray-Ransac. This algorithm combines the structural characteristics of three-dimensional lidar and uses ray segmentation to generate the original seed point set. The random sampling of Ransac algorithm is limited to the original seed point set, which reduces the probability that Ransac algorithm extracts outliers and reduces the calculation. The Ransac algorithm is used to modify the ground model parameters so that the algorithm can adapt to the undulating roads. The standard deviation of the distance from the point to the plane model is used as the distance threshold, and the allowable error range of the actual point cloud data is considered to effectively eliminate the abnormal points and error points. The algorithm was tested on the simulation platform and the test vehicle. The experimental results show that the lidar point cloud ground segmentation algorithm proposed in this paper takes an average of 5.784 milliseconds per frame, which has fast speed and good precision. It can adapt to uneven road surface and has high robustness.

scholarly journals 3D CNN with Visual Insights for Early Detection of Lung Cancer Using Gradient-Weighted Class Activation

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Eali Stephen Neal Joshua ◽  
Debnath Bhattacharyya ◽  
Midhun Chakkravarthy ◽  
Yung-Cheol Byun
Keyword(s):  
Low Dose ◽  
Binary Classification ◽  
Lung Nodule ◽  
Fine Grained ◽  
Internal Working ◽  
Nodule Detection ◽  
Classification Tasks ◽  
3D Cnn ◽  
Nodule Classification ◽  
Fold Cross Validation

The 3D convolutional neural network is able to make use of the full nonlinear 3D context information of lung nodule detection from the DICOM (Digital Imaging and Communications in Medicine) images, and the Gradient Class Activation has shown to be useful for tailoring classification tasks and localization interpretation for fine-grained features and visual explanation for the internal working. Gradient-weighted class activation plays a crucial role for clinicians and radiologists in terms of trusting and adopting the model. Practitioners not only rely on a model that can provide high precision but also really want to gain the respect of radiologists. So, in this paper, we explored the lung nodule classification using the improvised 3D AlexNet with lightweight architecture. Our network employed the full nature of the multiview network strategy. We have conducted the binary classification (benign and malignant) on computed tomography (CT) images from the LUNA 16 database conglomerate and database image resource initiative. The results obtained are through the 10-fold cross-validation. Experimental results have shown that the proposed lightweight architecture achieved a superior classification accuracy of 97.17% on LUNA 16 dataset when compared with existing classification algorithms and low-dose CT scan images as well.

scholarly journals Semantic Segmentation of 3D Point Cloud Based on Spatial Eight-Quadrant Kernel Convolution

2021 ◽  
Vol 13 (16) ◽  
pp. 3140
Author(s):  
Liman Liu ◽  
Jinjin Yu ◽  
Longyu Tan ◽  
Wanjuan Su ◽  
Lin Zhao ◽  
...  
Keyword(s):  
Point Cloud ◽  
Poor Performance ◽  
Semantic Segmentation ◽  
Point Clouds ◽  
Small Object ◽  
Fine Grained ◽  
3D Point Clouds ◽  
Kernel Convolution ◽  
Segmentation Accuracy ◽  
Indoor Scenes

In order to deal with the problem that some existing semantic segmentation networks for 3D point clouds generally have poor performance on small objects, a Spatial Eight-Quadrant Kernel Convolution (SEQKC) algorithm is proposed to enhance the ability of the network for extracting fine-grained features from 3D point clouds. As a result, the semantic segmentation accuracy of small objects in indoor scenes can be improved. To be specific, in the spherical space of the point cloud neighborhoods, a kernel point with attached weights is constructed in each octant, the distances between the kernel point and the points in its neighborhood are calculated, and the distance and the kernel points’ weights are used together to weight the point cloud features in the neighborhood space. In this case, the relationship between points are modeled, so that the local fine-grained features of the point clouds can be extracted by the SEQKC. Based on the SEQKC, we design a downsampling module for point clouds, and embed it into classical semantic segmentation networks (PointNet++, PointSIFT and PointConv) for semantic segmentation. Experimental results on benchmark dataset ScanNet V2 show that SEQKC-based PointNet++, PointSIFT and PointConv outperform the original networks about 1.35–2.12% in terms of MIoU, and they effectively improve the semantic segmentation performance of the networks for small objects of indoor scenes, e.g., the segmentation accuracy of small object “picture” is improved from 0.70% of PointNet++ to 10.37% of SEQKC-PointNet++.

scholarly journals Pointwise Rotation-Invariant Network with Adaptive Sampling and 3D Spherical Voxel Convolution

2020 ◽  
Vol 34 (07) ◽  
pp. 12717-12724
Author(s):  
Yang You ◽  
Yujing Lou ◽  
Qi Liu ◽  
Yu-Wing Tai ◽  
Lizhuang Ma ◽  
...  
Keyword(s):  
Adaptive Sampling ◽  
Point Cloud ◽  
Data Augmentation ◽  
Feature Matching ◽  
State Of The Art ◽  
Point Clouds ◽  
Rotation Invariant ◽  
Learning Framework ◽  
Point Set ◽  
Part Segmentation

Point cloud analysis without pose priors is very challenging in real applications, as the orientations of point clouds are often unknown. In this paper, we propose a brand new point-set learning framework PRIN, namely, Pointwise Rotation-Invariant Network, focusing on rotation-invariant feature extraction in point clouds analysis. We construct spherical signals by Density Aware Adaptive Sampling to deal with distorted point distributions in spherical space. In addition, we propose Spherical Voxel Convolution and Point Re-sampling to extract rotation-invariant features for each point. Our network can be applied to tasks ranging from object classification, part segmentation, to 3D feature matching and label alignment. We show that, on the dataset with randomly rotated point clouds, PRIN demonstrates better performance than state-of-the-art methods without any data augmentation. We also provide theoretical analysis for the rotation-invariance achieved by our methods.

scholarly journals Identification of Linear Vegetation Elements in a Rural Landscape Using LiDAR Point Clouds

2019 ◽  
Vol 11 (3) ◽  
pp. 292 ◽  
Author(s):  
Chris Lucas ◽  
Willem Bouten ◽  
Zsófia Koma ◽  
W. Kissling ◽  
Arie Seijmonsbergen
Keyword(s):  
Regional Planning ◽  
Point Cloud ◽  
Agricultural Land ◽  
3D Structure ◽  
Point Clouds ◽  
Bare Soil ◽  
Agricultural Landscapes ◽  
Rural Landscape ◽  
Conservation Strategies ◽  
Rural Landscapes

Modernization of agricultural land use across Europe is responsible for a substantial decline of linear vegetation elements such as tree lines, hedgerows, riparian vegetation, and green lanes. These linear objects have an important function for biodiversity, e.g., as ecological corridors and local habitats for many animal and plant species. Knowledge on their spatial distribution is therefore essential to support conservation strategies and regional planning in rural landscapes but detailed inventories of such linear objects are often lacking. Here, we propose a method to detect linear vegetation elements in agricultural landscapes using classification and segmentation of high-resolution Light Detection and Ranging (LiDAR) point data. To quantify the 3D structure of vegetation, we applied point cloud analysis to identify point-based and neighborhood-based features. As a preprocessing step, we removed planar surfaces such as grassland, bare soil, and water bodies from the point cloud using a feature that describes to what extent the points are scattered in the local neighborhood. We then applied a random forest classifier to separate the remaining points into vegetation and other. Subsequently, a rectangularity-based region growing algorithm allowed to segment the vegetation points into 2D rectangular objects, which were then classified into linear objects based on their elongatedness. We evaluated the accuracy of the linear objects against a manually delineated validation set. The results showed high user’s (0.80), producer’s (0.85), and total accuracies (0.90). These findings are a promising step towards testing our method in other regions and for upscaling it to broad spatial extents. This would allow producing detailed inventories of linear vegetation elements at regional and continental scales in support of biodiversity conservation and regional planning in agricultural and other rural landscapes.

Sign in / Sign up

Export Citation Format

Share Document