Advanced Concepts for Intelligent Vision Systems

This book constitutes the proceedings of the 20th INternational Conference on Advanced Concepts for Intelligent Vision Systems, ACIVS 2020, held in Auckland, New Zealand, in February 2020. The 48 papers presented in this volume were carefully reviewed and selected from a total of 78 submissions. They were organized in topical sections named: deep learning; biomedical image analysis; biometrics and identification; image analysis; image restauration, compression and watermarking; tracking, and mapping and scene analysis.

Inhalt

Deep Learning.- Desing data model for Big Data Analysis System.- Deep Learning-based Techniques for Plant Diseases Recognition in Real-Field Scenarios.- EpNet: a Deep Neural Network for Ear Detection in 3D Point Clouds.- Fire Segmentation in Still Images.- Region Proposal Oriented Approach for Domain Adaptive Object Detection.- Deep Convolutional Network-Based Framework for Melanoma Lesion Detection and Segmentation.- A Novel Framework for Early Fire Detection Using Terrestrial and Aerial 360-degree Images.- Biomedical Image Analysis.- Segmentation of Phase-Contrast MR Images for Aortic Pulse Wave Velocity Measurements.- On the Uncertainty of Retinal Artery-vein Classification with Dense Fully-convolutional Neural Networks.- Object Contour Refinement using Instance Segmentation in Dental Images.- Correction of Temperature Estimated from a Low-Cost Handheld Infrared Camera for Clinical Monitoring.- Bayesian Feature Pyramid Networks for Automatic Multi-Label Segmentation of Chest X-rays and Assessment of Cardio-Thoratic.- Deep-Learning for Tidemark Segmentation in Human Osteochondral Tissues Imaged with Micro-computed Tomography.- Quadratic Tensor Anisotropy Measures for Reliable Curvilinear Pattern Detection.- Biometrics and Identification.- Exposing Presentation Attacks by a Combination of Multi-intrinsic Image Properties, Convolutional Networks and Transfer Learning.- Multiview 3D Markerless Human Pose Estimation.- Clip-level Feature Aggregation: A Key Factor for Video-based Person Re-Identification.- Towards Approximating Personality Cues Through Simple Daily Activities.- Person Identification by Walking Gesture using Skeleton Sequences.- Verifying Kinship from RGB-D Face Data.- VA-StarGAN: Continuous Affect Generation.- Fast Iris Segmentation Algorithm for Visible Wavelength Images Based on Multi-Color Space.- A Local Flow Phase Stretch Transform for Robust Retinal Vessel Detection.- Evaluation of Unconditioned Deep Generative Synthesis of Retinal Images.- Image Analysis.- Dynamic Texture Representation Based on Hierarchical Local Patterns.- Temporal-clustering based Technique for Identifying Thermal Regions in Buildings.- Distance Weighted Loss for Forest Trail Detection using Semantic Line.- Localization of Map Changes by Exploiting SLAM Residuals.- Initial Pose Estimation of 3D Object with Severe Occlusion Using Deep Learning.- Automatic Focal Blur Segmentation based on Difference of Blur Feature using Theoretical Thresholding and Graphcuts.- Feature Map Augmentation to Improve Rotation Invariance in Convolutional Neural Networks.- Automatic Optical Inspection for Millimeter Scale Probe Surface Stripping Defects using Convolutional Neural Network.- Image restauration, Compression and Watermarking.- A New SVM-based Zero-watermarking Technique for3D Videos Traitor Tracing.- Design of Perspective Affine Motion Compensation for Versatile Video Coding (VVC).- Investigation of Coding Standards Performances on Optically Acquired and Synthetic Holograms.- Natural Images Enhancement Using Structure Extraction and Retinex.- Unsupervised Desmoking of Laparoscopy Images using Multi-scale DesmokeNet.- VLW-Net: A Very Light-Weight Convolutional Neural Network (CNN) for Single Image Dehazing.- An Improved GAN Semantic Image Inpainting.- Tracking, Mapping and Scene Analysis.- CUDA Implementation of a Point Cloud Shape Descriptor Method for Archaeological Studies.- Red-Green-Blue Augmented Reality Tags for Retail Stores.- Guided Stereo to Improve Depth Resolution of a Small Baseline Stereo Camera Using an Image Sequence.- SuperNCN: Neighbourhood Consensus Network for Robust Outdoor Scenes Matching.- Using Normal/ Abnormal Video Sequence Categorization to Efficient Facial Expression Recognition in the Wild.- Distributed Multi-Class Road User Tracking in Multi-Camera Network for Smart Traffic Applications.- Vehicles Tracking by combining Convolutional Neural Network based Segmentation and Optical Flow Estimation.- Real Time Embedded Person Detection and Tracking in Camera Streams.- Learning Target-Specific Response Attention for Siamese Network Based Visual Tracking.