This book constitutes the refereed proceedings of the 39th Computer Graphics International Conference on Advances in Computer Graphics, CGI 2022, held Virtually, during September 12-16, 2022.
The 45 full papers included in this book were carefully reviewed and selected from 139 submissions. They were organized in topical sections as follows: image analysis & processing; graphs & networks; estimation & feature matching; 3d reconstruction; rendering & animation; detection & recognition; colors, paintings & layout; synthesis & generation; ar & user interfaces; medical imaging; segmentation; object detection; image attention & perception; and modeling & simulation.
Inhaltsverzeichnis
IMAGE ANALYSIS & PROCESSING
. - Multi-granularity feature attention Fusion network for Image-Text Sentiment Analysis. - Toward efficient image denoising: A lightweight network with retargeting supervision driven knowledge distillation. - A pig pose estimation model for measuring pig s body size. - Topology-aware Learning for Semi-supervised Cross-domain Retinal Artery/Vein Classification. - Face Super-Resolution with Better Semantics and More Efficient Guidance. -
GRAPHS & NETWORKS
. - Layout and Display of Network Graphs on a Sphere. - Joint Matrix Factorization and Structure Preserving for Domain Adaptation. - Graph Adversarial Network with Bottleneck Adapter Tuning for Sign Language Production. -
ESTIMATION & FEATURE MATCHING
. - Facial Landmarks based Region-Level Data Augmentation for Gaze Estimation. - An Efficient Dense Depth Map Estimation Algorithm Using Direct Stereo Matching For Ultra-wide-angle Images. - Ad-RMS: Adaptive Regional Motion Statistics for Feature Matching Filtering. -
3D RECONSTRUCTION
. - Mobile Augmented Realty Based Visual Indoor Navigation. - Cost Volume Pyramid Network with Multi-strategies Range Searching for Multi-view Stereo. - Reconstructing the Surface Mesh Representation for Single Neuron. - WEmap: Weakness-enhancement mapping for 3D reconstruction with sparse image sequences. -
RENDERING & ANIMATION
. - Comparing Traditional Rendering Techniques to Deep Learning based Super-Resolution in Fire and Smoke Animations. - Real-Time Light Field Path Tracing. - Crowd Simulation with Detailed Body Motion and Interaction. - Towards Rendering the Style of 20th Century Cartoon Line Art in 3D Real-time. -
DETECTION & RECOGNITION
. - Face detection algorithm in classroom scene based on deep learning. - GRVT: Toward Effective Grocery Recognition via Vision Transformer. - A Transformer-based Cloth-irrelevant Patches Feature Extracting Method for Long-Term Cloth-Changing Person Re-identification. - Learning Unified Binary Feature Codes For Cross-illumination Palmprint Recognition. -
COLORS, PAINTINGS & LAYOUT
. - SemiPainter: Learning to draw semi-realistic paintings from the manga line drawings and flat shadow. - Hierarchical Bayesian Network Modeling and Layout of Huizhou Traditional Villages in Geographic Environment. - AE-GAN: Attention Embedded GAN For Irregular and Large-area Mask Face Image Inpainting. -
SYNTHESIS & GENERATION
. - Procedural Generation of Landscapes with Water Bodies Using Artificial Drainage Basins. - High-fidelity Dynamic Human Synthesis via UV-guided NeRF with Sparse Views. - Rec2Real: Semantics-Guided Photo-Realistic Image Synthesis Using Rough Urban Reconstruction Models. - 3D Digital City Structure Model Based on Image Modeling Technology. -
AR & USER INTERFACES
. - Augmented Reality-Based Home Interaction Layout and Evaluation. - LiteAR: A Framework to Estimate Lighting for Mixed Reality Sessions for Enhanced Realism. - Personalized User Interface Elements Recommendation System. -
MEDICAL IMAGING
. -
A feature point extraction method for capsule endoscope localization. - Automated Diagnosis Of Retinal Neovascularization Pathologies From Color Retinal Fundus Images. -
SEGMENTATION
. - DDCNet: A lightweight network with variable receptive field for real-time portrait segmentation in complex environment. - A chromosome segmentation method based on corner detection and watershed algorithm. - Voxel-based 3D shape segmentation using deep volumetric convolutional neural networks. -
OBJECT DETECTION
. - Few-shot detection based on an enhanced prototype for outdoor small forbidden objects. - Research on real-time forestry pest detection based on improved YOLOv5. - Power Line Detection Based on Feature Fusion Deep Learning Network. -
IMAGE ATTENTION & PERCEPTION
. - Wider and Higher: Intensive Integration and Global Foreground Perception for Image Matting. - Authenticity Identification of Qi Baishi s Shrimp Painting with Dynamic Token Enhanced Visual Transformer. -
MODELING & SIMULATION
. - An Optimized Material Point Method for Soil-Water Coupled Simulation. - SlimFliud-Net: Fast Fluid Simulation Using Admm Pruning.
