scholarly journals A 3D Keypoints Voting Network for 6DoF Pose Estimation in Indoor Scene

Machines ◽  
2021 ◽  
Vol 9 (10) ◽  
pp. 230
Author(s):  
Huikai Liu ◽  
Gaorui Liu ◽  
Yue Zhang ◽  
Linjian Lei ◽  
Hui Xie ◽  
...  
Keyword(s):  
Pose Estimation ◽  
Selection Method ◽  
Feature Representation ◽  
Experimental Results ◽  
Two Stage ◽  
Feature Map ◽  
Indoor Scene ◽  
Salient Regions ◽  
Split Attention ◽  
Farthest Point

This paper addresses the problem of instance-level 6DoF pose estimation from a single RGBD image in an indoor scene. Many recent works have shown that a two-stage network, which first detects the keypoints and then regresses the keypoints for 6d pose estimation, achieves remarkable performance. However, the previous methods concern little about channel-wise attention and the keypoints are not selected by comprehensive use of RGBD information, which limits the performance of the network. To enhance RGB feature representation ability, a modular Split-Attention block that enables attention across feature-map groups is proposed. In addition, by combining the Oriented FAST and Rotated BRIEF (ORB) keypoints and the Farthest Point Sample (FPS) algorithm, a simple but effective keypoint selection method named ORB-FPS is presented to avoid the keypoints appear on the non-salient regions. The proposed algorithm is tested on the Linemod and the YCB-Video dataset, the experimental results demonstrate that our method outperforms the current approaches, achieves ADD(S) accuracy of 94.5% on the Linemod dataset and 91.4% on the YCB-Video dataset.

A Configurable Tilt License Plate Correction Method Based on Parallel Lines

2013 ◽  
Vol 385-386 ◽  
pp. 1429-1433 ◽  
Author(s):  
Zhong Yan Liang ◽  
San Yuan Zhang
Keyword(s):  
Recognition System ◽  
Correction Method ◽  
Selection Method ◽  
Experimental Results ◽  
License Plate ◽  
License Plate Recognition ◽  
Parallel Lines ◽  
Bounding Box ◽  
Bounding Boxes ◽  
Straight Lines

The tilt license plate correction is an important part of the license plate recognition system. Traditional correction methods are based on one theory. It is difficult to use the advantages of different approaches. We propose some methods to help improve the tile license plate correction: a bounding box selection method based on similar height and a mutual correction method based on fitted parallel straight lines. Moreover, we use wide bounding boxes to segment touched characters. If the method based on parallel lines fails, another method, such as PCA-based one, can be used for complement. Experimental results show the proposed method outperforms others.

Local Graph Edge Partitioning

2021 ◽  
Vol 12 (5) ◽  
pp. 1-25
Author(s):  
Shengwei Ji ◽  
Chenyang Bu ◽  
Lei Li ◽  
Xindong Wu
Keyword(s):  
Real World ◽  
Graph Partitioning ◽  
Large Scale ◽  
Complete Information ◽  
Local Information ◽  
Experimental Results ◽  
Two Stage ◽  
Graph Computation ◽  
Local Graph ◽  
Edge Partitioning

Graph edge partitioning, which is essential for the efficiency of distributed graph computation systems, divides a graph into several balanced partitions within a given size to minimize the number of vertices to be cut. Existing graph partitioning models can be classified into two categories: offline and streaming graph partitioning models. The former requires global graph information during the partitioning, which is expensive in terms of time and memory for large-scale graphs. The latter creates partitions based solely on the received graph information. However, the streaming model may result in a lower partitioning quality compared with the offline model. Therefore, this study introduces a Local Graph Edge Partitioning model, which considers only the local information (i.e., a portion of a graph instead of the entire graph) during the partitioning. Considering only the local graph information is meaningful because acquiring complete information for large-scale graphs is expensive. Based on the Local Graph Edge Partitioning model, two local graph edge partitioning algorithms—Two-stage Local Partitioning and Adaptive Local Partitioning—are given. Experimental results obtained on 14 real-world graphs demonstrate that the proposed algorithms outperform rival algorithms in most tested cases. Furthermore, the proposed algorithms are proven to significantly improve the efficiency of the real graph computation system GraphX.

scholarly journals A Novel Detector Based on Convolution Neural Networks for Multiscale SAR Ship Detection in Complex Background

Sensors ◽  
2020 ◽  
Vol 20 (9) ◽  
pp. 2547 ◽  
Author(s):  
Wenxin Dai ◽  
Yuqing Mao ◽  
Rongao Yuan ◽  
Yijing Liu ◽  
Xuemei Pu ◽  
...  
Keyword(s):  
Region Of Interest ◽  
Feature Representation ◽  
Feature Maps ◽  
Sar Images ◽  
Feature Map ◽  
The Public ◽  
Single Feature ◽  
Great Performance ◽  
Residual Block ◽  
Fusion Feature

Convolution neural network (CNN)-based detectors have shown great performance on ship detections of synthetic aperture radar (SAR) images. However, the performance of current models has not been satisfactory enough for detecting multiscale ships and small-size ones in front of complex backgrounds. To address the problem, we propose a novel SAR ship detector based on CNN, which consist of three subnetworks: the Fusion Feature Extractor Network (FFEN), Region Proposal Network (RPN), and Refine Detection Network (RDN). Instead of using a single feature map, we fuse feature maps in bottom–up and top–down ways and generate proposals from each fused feature map in FFEN. Furthermore, we further merge features generated by the region-of-interest (RoI) pooling layer in RDN. Based on the feature representation strategy, the CNN framework constructed can significantly enhance the location and semantics information for the multiscale ships, in particular for the small ships. On the other hand, the residual block is introduced to increase the network depth, through which the detection precision could be further improved. The public SAR ship dataset (SSDD) and China Gaofen-3 satellite SAR image are used to validate the proposed method. Our method shows excellent performance for detecting the multiscale and small-size ships with respect to some competitive models and exhibits high potential in practical application.

Improving Self-interacting Proteins Prediction Accuracy Using Protein Evolutionary Information and Weighed-Extreme Learning Machine

2019 ◽  
Vol 14 (2) ◽  
pp. 115-122 ◽  
Author(s):  
Ji-Yong An ◽  
Yong Zhou ◽  
Lei Zhang ◽  
Qiang Niu ◽  
Da-Fu Wang
Keyword(s):  
Extreme Learning Machine ◽  
Functional Organization ◽  
Web Server ◽  
Feature Representation ◽  
Experimental Results ◽  
Evolutionary Information ◽  
Svm Classifier ◽  
Interacting Proteins ◽  
Learning Machine ◽  
Human Dataset

Background: Self Interacting Proteins (SIPs) play an essential role in various aspects of the structural and functional organization of the cell. Objective: In the study, we presented a novelty sequence-based computational approach for predicting Self-interacting proteins using Weighed-Extreme Learning Machine (WELM) model combined with an Autocorrelation (AC) descriptor protein feature representation. Method: The major advantage of the proposed method mainly lies in adopting an effective feature extraction method to represent candidate self-interacting proteins by using the evolutionary information embedded in PSI-BLAST-constructed Position Specific Scoring Matrix (PSSM); and then employing a reliable and effective WELM classifier to perform classify. </P><P> Result: In order to evaluate the performance, the proposed approach is applied to yeast and human SIP datasets. The experimental results show that our method obtained 93.43% and 98.15% prediction accuracies on yeast and human dataset, respectively. Extensive experiments are carried out to compare our approach with the SVM classifier and existing sequence-based method on yeast and human dataset. Experimental results show that the performance of our method is better than several other state-of-theart methods. Conclusion: It is demonstrated that the proposed method is suitable for SIPs detection and can execute incredibly well for identifying Sips. In order to facilitate extensive studies for future proteomics research, we developed a freely available web server called WELM-AC-SIPs in Hypertext Preprocessor (PHP) for predicting SIPs. The web server including source code and the datasets are available at http://219.219.62.123:8888/WELMAC/.

scholarly journals End-to-End Game-Focused Learning of Adversary Behavior in Security Games

2020 ◽  
Vol 34 (02) ◽  
pp. 1378-1386
Author(s):  
Andrew Perrault ◽  
Bryan Wilder ◽  
Eric Ewing ◽  
Aditya Mate ◽  
Bistra Dilkina ◽  
...  
Keyword(s):  
Expected Utility ◽  
Optimization Problem ◽  
Predictive Accuracy ◽  
Experimental Results ◽  
Limited Data ◽  
Two Stage ◽  
Security Games ◽  
End To End ◽  
Feature Values ◽  
Adversary Model

Stackelberg security games are a critical tool for maximizing the utility of limited defense resources to protect important targets from an intelligent adversary. Motivated by green security, where the defender may only observe an adversary's response to defense on a limited set of targets, we study the problem of learning a defense that generalizes well to a new set of targets with novel feature values and combinations. Traditionally, this problem has been addressed via a two-stage approach where an adversary model is trained to maximize predictive accuracy without considering the defender's optimization problem. We develop an end-to-end game-focused approach, where the adversary model is trained to maximize a surrogate for the defender's expected utility. We show both in theory and experimental results that our game-focused approach achieves higher defender expected utility than the two-stage alternative when there is limited data.

scholarly journals How to manipulate droplet jetting from needle type jet dispensers

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Thanh Huy Phung ◽  
Kye-Si Kwon
Keyword(s):  
Selection Method ◽  
Parameter Selection ◽  
Measurement Methods ◽  
Travel Distance ◽  
Experimental Results ◽  
Effective Parameter ◽  
Single Droplet ◽  
Optimal Range ◽  
Needle Type ◽  
Bounce Back

AbstractThe needle-type inkjet dispenser has been widely used for various research and industrial purposes. The droplet jetting from the dispenser is closely related to the needle motion, which strikes against the nozzle seat. The strike of the needle on the nozzle seat often cause additional impact due to the bounce back, which may produce multiple droplets per jetting trigger. However, the needle motion is difficult to measure, and the actual behaviors have been known little. In this study, we measured the needle motion using an accelerometer and visualized jetting images to understand jetting behavior in relation to the needle motion. Then, we investigated various parameter effects on needle motion and jetting behaviors based on our proposed measurement methods. From the experimental results, we found that needle travel distance should be in the optimal range in order to produce single droplet per jetting trigger. In conclusion, we proposed an effective parameter selection method for the optimal jetting based on understanding of the jetting physics.

Research on Pose Estimation for Electric Vehicle’s Fast-Swap Battery Box Based on Visual Information

2012 ◽  
Vol 253-255 ◽  
pp. 2252-2257
Author(s):  
Yu Ming Wu ◽  
Shuo Liu ◽  
Gao Yang Zhang ◽  
Xiao Yan Yin ◽  
Ming Yu Zhao ◽  
...  
Keyword(s):  
Coordinate System ◽  
Pose Estimation ◽  
Visual Information ◽  
Experimental Results ◽  
Noise Effects ◽  
Simulation Results ◽  
Camera Coordinate System

For the reason of difficult to get battery box pose information, we research the battery box pose measure method based on visual information. We get the coplanar four points at the lines constraints which extracted form image. We get the pose relationship between the battery box coordinate system and camera coordinate system, and then calculate the average of the measure results to reduce noise effects for measure precision. Simulation results show that the method through calculate average of the measure results can effectively reduce noise effects for measure accuracy. The actual experimental results show that the pose estimate accuracy is meet robot requirements for battery swap.

Sign in / Sign up

Export Citation Format

Share Document