LEADER 10735nam 2200505 450 001 996464413603316 005 20231110221651.0 010 $a981-16-7189-3 035 $a(CKB)5470000001298855 035 $a(MiAaPQ)EBC6795914 035 $a(Au-PeEL)EBL6795914 035 $a(OCoLC)1281772636 035 $a(PPN)258297603 035 $a(EXLCZ)995470000001298855 100 $a20220720d2021 uy 0 101 0 $aeng 135 $aurcnu|||||||| 181 $ctxt$2rdacontent 182 $cc$2rdamedia 183 $acr$2rdacarrier 200 00$aImage and graphics technologies and applications $e16th Chinese conference on image and graphics technologies, IGTA 2021, Beijing, China, June 6-7, 2021, revised selected papers /$fedited by Yongtian Wang and Weitao Song 210 1$aSingapore :$cSpringer,$d[2021] 210 4$d©2021 215 $a1 online resource (283 pages) 225 1 $aCommunications in Computer and Information Science ;$vv.1480 311 $a981-16-7188-5 320 $aIncludes bibliographical references and index. 327 $aIntro -- Preface -- Organization -- Contents -- Image Processing and Enhancement Techniques -- Residual Multi-resolution Network for Hyperspectral Image Denoising -- 1 Introduction -- 2 Methodology -- 2.1 Network Architecture -- 2.2 Residual Multi-Resolution Block -- 3 Experimental Results and Analysis -- 3.1 Experimental Setup -- 3.2 Results and Analysis -- 4 Conclusion -- References -- Skin Reflectance Reconstruction Based on the Polynomial Regression Model -- 1 Introduction -- 2 Related Work -- 2.1 Human Skin Spectral -- 2.2 Polynomial Regression Model -- 3 Algorithms -- 3.1 The Xiao's Algorithm -- 3.2 The Proposed Algorithm -- 3.3 Regularization -- 4 Experiment Setup -- 4.1 Evaluation Metrics -- 4.2 Experiment Data -- 5 Result and Discussion -- 6 Conclusion -- References -- From Deep Image Decomposition to Single Depth Image Super-Resolution -- 1 Introduction -- 2 The Proposed Method -- 2.1 Depth Dual Decomposition Block -- 2.2 Depth Image Initialization Block -- 2.3 Depth Image Rebuilding Block -- 2.4 Loss Function -- 3 Experimental Results -- 3.1 Training Details -- 3.2 The Objective Quality Comparison of Different Methods -- 3.3 The Visual Quality Comparison of Different Methods -- 4 Conclusion -- References -- Classification of Solar Radio Spectrum Based on VGG16 Transfer Learning -- 1 Introduction -- 2 VGG16 and Transfer Learning -- 2.1 VGG16 Convolutional Neural Network Model -- 2.2 Transfer Learning -- 3 Spectrum Image Preprocessing and Classification Algorithm -- 3.1 Preprocessing of Solar Radio Spectrum Data -- 3.2 VGG16 Transfer Learning Algorithm -- 4 Experimental Results and Analysis -- 5 Conclusion -- References -- A Channel Attention-Based Convolutional Neural Network for Intra Chroma Prediction of H.266 -- 1 Introduction -- 2 Proposed Method -- 2.1 Neighboring Block Information Extraction Module. 327 $a2.2 Current Block Information Extraction Module -- 2.3 Feature Fusion Module -- 3 Experiment Results -- 3.1 The Robustness Analysis of Chroma Prediction Performance -- 3.2 Coding Performance -- 4 Conclusion -- References -- Biometric Identification Techniques -- A Novel Deep Residual Attention Network for Face Mosaic Removal -- 1 Introduction -- 2 The Proposed Method -- 2.1 Parallel Residual Block -- 2.2 Channel Attention -- 2.3 Pixel Attention -- 3 Experimental Results -- 3.1 Experimental Settings -- 3.2 Quantitative and Qualitative Evaluation -- 4 Conclusion -- References -- Machine Vision and 3D Reconstruction -- Pretrained Self-supervised Material Reflectance Estimation Based on a Differentiable Image-Based Renderer -- 1 Introduction -- 2 Related Work -- 3 Self-supervised Architectures -- 4 Image-Based Differentiable Renderer -- 4.1 Normal Representation -- 4.2 Ambient Light Representation -- 4.3 Material Representation -- 5 Synthetic Dataset -- 6 Loss Functions -- 7 Experiments -- 8 Conclusion -- References -- Image/Video Big Data Analysis and Understanding -- Object-Aware Attention in Few-Shot Learning -- 1 Introduction -- 2 Related Work -- 2.1 Few-Shot Learning -- 2.2 Saliency Object Detection -- 3 Method -- 3.1 Preliminaries -- 3.2 Framework -- 3.3 Object-Aware Attention Module -- 4 Experiments -- 4.1 Datasets -- 4.2 Baseline Models and Experiment Details -- 4.3 Experiment Result -- 5 Conclusion -- References -- Simultaneously Predicting Video Object Segmentation and Optical Flow Without Motion Annotations -- 1 Introduction -- 2 Methodology -- 2.1 Base Model -- 2.2 Object Flow Synthesizing -- 3 Network Implementation and Training -- 3.1 Offline Training -- 3.2 Online Training -- 4 Experimental Results -- 4.1 Dataset and Evaluation Metrics -- 4.2 Ablation Study on Video Object Segmentation -- 4.3 Segmentation Results -- 4.4 Object Flow. 327 $a4.5 Runtime Analysis -- 5 Concluding Remarks -- References -- Recognition of Bending Deformed Pipe Sections in Geological Disaster Area Based on an Ensemble Learning Model -- 1 Introduction -- 2 Feature Engineering -- 2.1 Typical Pipe Section -- 2.2 IMU Data Feature Construction -- 3 Ensemble Learning -- 3.1 Support Vector Machine -- 3.2 K-means Clustering -- 3.3 Ensemble Learning Based on Voting Method -- 3.4 Analysis of Results -- 4 Conclusion -- References -- Memory Bank Clustering for Self-supervised Contrastive Learning -- 1 Introduction -- 2 Related Work -- 2.1 Self Supervised Contrastive Learning -- 2.2 Memory Bank Based Method -- 3 Method -- 3.1 Contrastive Learning with Memory Bank Clustering -- 3.2 Training with Dynamic Memory Bank -- 3.3 Contrastive Loss with Dynamic Memory Bank -- 4 Experiment -- 4.1 Experiment Details -- 4.2 Experimental Results -- 5 Conclusion -- References -- Robust Visual Question Answering Based on Counterfactual Samples and Relationship Perception -- 1 Introduction -- 2 Related Work -- 2.1 Visual Question Answering -- 2.2 Counterfactual Sample Synthesis Mechanism for VQA -- 2.3 Relationship Perception in VQA -- 3 Method -- 3.1 Relationship Perception Network -- 3.2 Counterfactual Sample Generation Mechanism -- 3.3 Loss Functions -- 4 Experiment -- 4.1 Dataset -- 4.2 Parameter Setting -- 4.3 Performance on the VQA-CP v2 Dataset -- 4.4 Performance on the VQA V2 Dataset -- 4.5 Ablation Experiment of CSRP Model -- 5 Conclusion -- References -- Prototype Generation Based Shift Graph Convolutional Network for Semi-supervised Anomaly Detection -- 1 Introduction -- 2 Related Work -- 2.1 Video Anomaly Detection -- 2.2 Graph Convolutional Networks -- 2.3 Prototype Generation Module -- 3 Method -- 3.1 Architecture Overview -- 3.2 Feature Extraction Module -- 3.3 Prototype Generation Module -- 3.4 VAD Objective Functions. 327 $a4 Experiments -- 4.1 Experiments on ShanghaiTech -- 4.2 Ablation Study -- 5 Conclusion -- References -- Computer Graphics -- A New Image Super-Resolution Reconstruction Algorithm Based on Hybrid Diffusion Model -- 1 Introduction -- 2 Related Work -- 2.1 Characteristics of Fractional Differential Operators -- 2.2 Fractional Anisotropic Diffusion Model -- 3 Model Algorithm and Design -- 3.1 Proposal of Mixed Diffusion Model -- 3.2 Realization of Adaptive Diffusion Function -- 3.3 Numerical Calculation of the Algorithm -- 4 Experiment and Analysis -- 5 Conclusion -- References -- Research on Global Contrast Calculation Considering Color Differences -- 1 Introduction -- 2 Related Work -- 3 Method -- 3.1 Traditional Method of Global Contrast -- 3.2 Global Contrast Calculation Considering Color Differences -- 4 Experimental Results -- 5 Conclusion -- References -- Virtual Reality and Human-Computer Interaction -- Research on Key Technologies and Function Analysis of Live Interactive Classroom in AI+ Era -- 1 Introduction -- 2 Advantages and Disadvantages of Webcast Classroom -- 2.1 Advantages of Webcast Class -- 2.2 Disadvantages of Webcast Class -- 3 AI Plus Key Technology of Synchronous Live Broadcast in Class -- 3.1 Face Recognition Technology -- 3.2 Human Behavior and Expression Analysis -- 3.3 Natural Language Processing -- 3.4 Knowledge Graph and Expert Systems -- 4 The Main Functions of Open Education AI Live Class -- 4.1 Basic Framework of the Platform -- 4.2 Introduction of AI Live-Streaming Classroom Functions -- 5 Conclusion -- References -- Applications of Image and Graphics -- BrainSeg R-CNN for Brain Tumor Segmentation -- 1 Introduction -- 2 Method -- 2.1 Mask R-CNN -- 2.2 BrainSeg R-CNN -- 3 Experiments -- 3.1 Dataset and Settings -- 3.2 Compared Experiments Using Slices with Tumors -- 3.3 Compared Experiments Using Whole Brain Image. 327 $a4 Conclusion -- References -- A Real-Time Tracking Method for Satellite Video Based on Long-Term Tracking Framework -- 1 Introduction -- 2 Method Framework -- 2.1 Feature Extraction -- 2.2 Re-detection Mechanism -- 2.3 PCA Based Dimensionality Reduction -- 3 Experimental Results and Analysis -- 4 Conclusion -- References -- Fourier Series Fitting of Space Object Orbit Data -- 1 Introduction -- 2 Related Works -- 2.1 TLE Orbit Data -- 2.2 Orbit Elements -- 3 Data Fitting -- 3.1 Commonly Used Fitting Algorithms -- 3.2 Fourier Series Fitting -- 4 Experiment and Analysis -- 4.1 Classification Selection of Experimental Samples -- 4.2 Data Pre-processing -- 4.3 Evaluation Criteria of Fitting Results -- 4.4 Data Fitting -- 4.5 Predicting Results -- 5 Conclusion -- References -- Mapping Methods in Teleoperation of the Mars Rover -- 1 Introduction -- 2 Mars Rover Teleoperation Control Mode -- 3 Multi-scale Mapping Mode Based on Multi-source Data -- 3.1 Large Area Mapping Using Orbit Images Based on the Two-Stage Method -- 3.2 Landing Site Mapping Using Descent Sequence Images -- 3.3 3D Terrain Reconstruction Based on Image Fusion -- 3.4 Wide Baseline Mapping Using Multi-site Images -- 4 Conclusions -- References -- Other Research Works and Surveys Related to the Applications of Image and Graphics Technology -- A Regularized Limited-Angle CT Reconstruction Model Based on Sparse Multi-level Information Groups of the Images -- 1 Introduction -- 2 Analysis of Group Sparsity Model -- 3 The Proposed Model -- 4 Algorithms -- 5 Numerical Experiments -- 6 Conclusions and Prospects -- References -- Author Index. 410 0$aCommunications in Computer and Information Science 606 $aSemantics 615 0$aSemantics. 676 $a636.005 702 $aWang$b Yongtian 702 $aSong$b Weitao 801 0$bMiAaPQ 801 1$bMiAaPQ 801 2$bMiAaPQ 906 $aBOOK 912 $a996464413603316 996 $aImage and Graphics Technologies and Applications$92568128 997 $aUNISA