| |
|
|
|
|
|
|
|
|
1. |
Record Nr. |
UNISA996464496203316 |
|
|
Titolo |
Image and graphics : 11th International Conference, ICIG 2021, Haikou, China, August 6-8, 2021, Proceedings, Part II / / edited by Yuxin Peng [and five others] |
|
|
|
|
|
|
|
Pubbl/distr/stampa |
|
|
Cham, Switzerland : , : Springer, , [2021] |
|
©2021 |
|
|
|
|
|
|
|
|
|
ISBN |
|
|
|
|
|
|
Descrizione fisica |
|
1 online resource (849 pages) |
|
|
|
|
|
|
Collana |
|
Lecture Notes in Computer Science ; ; v.12889 |
|
|
|
|
|
|
Disciplina |
|
|
|
|
|
|
Soggetti |
|
Computer graphics |
Image processing - Digital techniques |
|
|
|
|
|
|
|
|
Lingua di pubblicazione |
|
|
|
|
|
|
Formato |
Materiale a stampa |
|
|
|
|
|
Livello bibliografico |
Monografia |
|
|
|
|
|
Note generali |
|
|
|
|
|
|
Nota di contenuto |
|
Intro -- Preface -- Organization -- Contents - Part II -- Artificial Intelligence -- A Calibration Strategy for Smart Welding -- 1 Introduction -- 2 Background -- 3 Calibration Method and Models -- 3.1 Camera Calibration -- 3.2 Laer Calibration -- 3.3 Hand-Eye Calibration -- 4 Experiment and Results -- 4.1 Image Processing -- 4.2 Transformation Image Coordinates to Camera Co-ordinate -- 4.3 Transformation Camera Coordinates to Robot Coordinate -- 5 Conclusion -- 6 Future Work -- References -- Brain Connectivity: Exploring from a High-Level Topological Perspective -- 1 Introduction -- 2 Materials and Methods -- 2.1 Participants and Data Acquisition -- 2.2 Data Preprocessing -- 2.3 Subnetwork Embedding -- 2.4 SE-Based Identification -- 3 Results -- 3.1 The Stable Relationship Between SEs -- 3.2 Traditional Relationship Preserved in SE Vectors -- 3.3 Comparison Between the Traditional and Reconstructed Coupling -- 3.4 Coupling in Schizophrenia -- 4 Discussion -- 4.1 A Subsection Sample -- 5 Conclusion -- References -- Hetero-STAN: Crowd Flow Prediction by Heterogeneous Spatio-Temporal Attention Network -- 1 Introduction -- 2 Related Works -- 3 Methodology -- 3.1 Formulation of Spatio-Temporal Prediction -- 3.2 Heterogeneous Feature Enhancement -- 3.3 Spatio-Temporal Attention Mechanism -- 4 Experiments and Analyses -- 5 Conclusion -- References -- UAV Track |
|
|
|
|
|
|
|
|
|
Planning Algorithm Based on Graph Attention Network and Deep Q Network -- 1 Introduction -- 2 Track Planning Model Based on MDP -- 3 UAV Track Planning System Based on DQN -- 3.1 Basic Elements of Reinforcement Learning -- 3.2 Deep Q Network -- 3.3 System Model of Track Planning Based on DQN -- 4 Flight State Optimization Based on CNN and GAT -- 4.1 Feature Extraction Based on Pre-trained ResNet -- 4.2 State Optimization Model Based on Graph Attention Network. |
4.3 Track Planning System Based on Flight State Optimization -- 5 Simulation and Discussion -- 5.1 Evaluation Index -- 5.2 Training and Testing -- 6 Conclusion -- References -- Object 6DoF Pose Estimation for Power Grid Manipulating Robots -- 1 Introduction -- 2 6DoF Pose Estimation -- 2.1 Pre-processing Stage -- 2.2 Instance Segmentation -- 2.3 Keypoints Prediction -- 2.4 2D-to-3D Projection -- 3 Experiments -- 3.1 Design the Clamp's Dataset -- 3.2 Results on Public Dataset Linemod -- 3.3 Results on Power Manipulating Robot -- 4 Conclusion -- References -- Novel Augmented Reality System for Oral and Maxillofacial Surgery -- 1 Introduction -- 2 Experimental Setup -- 3 Methods -- 3.1 System Framework -- 3.2 Virtual Model Production -- 3.3 Marker Recognition -- 3.4 Fusion View Generation -- 4 Experiments and Results -- 4.1 Experiment on Skull Phantom -- 4.2 Cadaveric Mandible Experiment -- 5 Conclusion -- References -- A New Dataset and Recognition for Egocentric Microgesture Designed by Ergonomists -- 1 Introduction -- 2 Realated Work -- 2.1 The Design of Gestures -- 2.2 Gesture Recognition -- 3 Methodology -- 3.1 Microgesture Set -- 3.2 Microgesture Dataset -- 3.3 Microgesture Recognition -- 4 Results -- 5 Conclusion -- References -- Machine Learning Architectures and Formulations -- Orthogonal Dual Graph Regularized Nonnegative Matrix Factorization -- 1 Introduction -- 2 Related Works -- 2.1 NMF -- 2.2 ONMF -- 2.3 GNMF -- 2.4 DNMF -- 3 Proposed Methods -- 3.1 ODNMF-U -- 3.2 ODNMF-V -- 3.3 ODNMF -- 4 Experimental Results -- 4.1 Experimental Setting -- 4.2 Datasets Description -- 4.3 Results Discussion -- 5 Conclusions -- References -- Robust Recovery of Low Rank Matrix by Nonconvex Rank Regularization -- 1 Introduction -- 2 Problem Formulation and Its Optimization -- 3 Convergence Analysis -- 4 Numerical Results -- 5 Conclusions -- References. |
Free Adversarial Training with Layerwise Heuristic Learning -- 1 Introduction -- 2 Related Work -- 2.1 Adversarial Defense -- 2.2 Adversarial Training -- 2.3 Training Acceleration -- 2.4 Adversarial Training Acceleration -- 3 Method -- 3.1 Adversarial Training -- 4 Experiments -- 4.1 CIFAR-10 Experiments -- 4.2 ImageNet Experiments -- 5 Conclusion -- References -- Adversarial Attack with KD-Tree Searching on Training Set -- 1 Introduction -- 2 Related Work -- 2.1 Fast Gradient Sign Method (FGSM) -- 2.2 Optimization-Based Attack (C& -- W) -- 2.3 Hot/Code Attack -- 2.4 Boundary Attack -- 3 Method -- 4 Experiments -- 4.1 Dataset Used in the Experiment -- 4.2 The Result of the Experiment -- 4.3 Performance Evaluation -- 5 Conclusions -- References -- Mixup Without Hesitation -- 1 Introduction -- 2 Related Work -- 3 Mixup Without Hesitation (mWh) -- 3.1 Observations -- 3.2 Algorithm -- 4 Experiments -- 4.1 Experiments on Image Classification Tasks -- 4.2 Transferability in CutMix -- 5 Discussion and Conclusion -- References -- An LDA and RBF-SVM Based Classification Method for Inertinite Macerals of Coal -- 1 Introduction -- 2 The Selection of Initial Features -- 2.1 Characteristic Analysis of Microscopic Image of Inertinite -- 2.2 Texture Feature Extraction -- 2.3 Gray Level Distribution Based Feature Extraction -- 3 Feature Extraction with LDA -- 4 Method of Classification -- 4.1 Construction of Classifier -- 4.2 Selection of |
|
|
|
|
|
|
|
Kernel Function -- 5 Experiment and Result Discussion -- 5.1 Implementation of Algorithm -- 5.2 Experimental Results and Discussion -- 6 Conclusion -- References -- A Weighted Non-convex Restoration Model for Hyperspectral Image -- 1 Introduction -- 2 Related Works and Preliminaries -- 3 A Weighted Non-convex Low-Rank Model for HSI Restoration -- 4 Experiments and Discussion -- 4.1 Quantitative Comparison. |
4.2 Visual Quality Comparison -- 5 Conclusion -- References -- Edge-Based Blur Kernel Estimation Using Sparse Representation and Self-similarity -- 1 Introduction -- 2 Sparse Representation and Multi-scale Self-Similarity -- 2.1 Sparse Representation -- 2.2 Multi-scale Self-similarity and Non-local Regularization -- 3 Blind Deconvolution -- 3.1 Use of Cross-scale Self-similarity -- 3.2 Analysis on Regularizers -- 3.3 Modeling and Optimization -- 3.4 Implementation -- 4 Experiments -- 4.1 Quantitative Evaluation with Reference to Ground Truth -- 4.2 Qualitative Comparison on Real Images -- 5 Conclusion -- References -- Normal Distribution Function on Descriptor Extraction -- 1 Introduction -- 2 Related Work -- 3 Method -- 3.1 Triplet Loss Function -- 3.2 Topological Relationship with Normal Distribution -- 4 Experiments -- 4.1 UBC Photo Tourism -- 4.2 HPatches -- 4.3 Wide Baseline Stereo (W1BS) -- 5 Conclusion -- References -- Supervised Non-negative Matrix Factorization Induced by Huber Loss -- 1 Introduction -- 2 Proposed HSNMF -- 2.1 HSNMF Model -- 2.2 Update Rules of HSNMF -- 2.3 Convergence of the Proposed HSNMF Algorithm -- 2.4 HSNMF Algorithm -- 3 Experimental Results -- 3.1 Results on Non-noisy Images -- 3.2 Results on Noisy Images -- 4 Conclusions -- References -- Small Sample Identification for Specific Emitter Based on Adversarial Embedded Networks -- 1 Introduction -- 2 Formation of the Time-Frequency Spectrogram -- 3 Construction of Adversarial Embedded Network Model -- 3.1 Generative Model -- 3.2 Discriminative Model -- 3.3 Classifier -- 3.4 Objective Function and Loss Function -- 4 Radar Emitters Recognition -- 5 Communication Station Emitters Recognition -- 6 Conclusion -- References -- Multi-mode Tensor Singular Value Decomposition for Low-Rank Image Recovery -- 1 Introduction -- 2 Notations and Preliminaries -- 3 Method. |
3.1 The Proposed Multi-mode Tensor Singular Value Decomposition -- 3.2 The Related Convex Optimization Works -- 3.3 The Proposed Multi-dimensional Square Model via TNN for Tensor Completion -- 4 Applications -- 4.1 Color Image Completion -- 4.2 Multispectral Image Completion -- 5 Conclusions -- References -- Hyperspectral Anomaly Detection Based on Tensor Truncated Nuclear Norm and Linear Total Variation Regularization -- 1 Introduction -- 2 Tensor Notations -- 3 Proposed Methods -- 3.1 TNNLTV for Anomaly Detection -- 3.2 ADMM Optimization Algorithm -- 4 Experiment -- 4.1 Data Sets Description -- 4.2 Performance -- 5 Conclusion -- References -- Face, Gesture, and Body Pose -- Deep Human Pose Estimation via Self-guided Learning -- 1 Introduction -- 2 Related Work -- 2.1 Self-Ensemble -- 2.2 Knowledge Distilling -- 3 Proposed Self-guided Learning -- 3.1 Problem Formulation -- 3.2 Model-Temporal Ensemble Guidance Learning -- 3.3 Self-guided Joint Loss Design -- 3.4 Overall Framework -- 4 Experimental Results -- 4.1 Implementation Details -- 4.2 Evaluation Metric -- 4.3 Overall Performance -- 4.4 Ablation Study -- 5 Conclusion -- References -- Binary Convolutional Neural Networks for Facial Action Unit Detection -- 1 Introduction -- 2 Related Work -- 3 Proposed Method -- 3.1 Fine-Tuned ResNet-18 -- 3.2 MDFS: Multi-layer Decision Fusion Strategy -- 3.3 BMDFS: MDFS with Binary Weights -- 4 Experiments -- 4.1 Datasets and Settings -- |
|
|
|
|
|
|
|
|
4.2 Experimental Results -- 5 Conclusion -- References -- Efficient Example Mining for Anchor-Free Face Detection -- 1 Introduction -- 2 Related Work -- 2.1 Anchor Free Detection -- 2.2 Label Assignment -- 3 Efficient Example Mining -- 3.1 Pipeline -- 3.2 Density Balance Scale Assignment -- 3.3 Multi-scale ERF Prior -- 4 Experiments -- 4.1 Dataset -- 4.2 Implementation Details -- 5 Conclusion -- References. |
6D Object Pose Estimation with Mutual Attention Fusion. |
|
|
|
|
|
| |