Advances in Visual Computing

This two-volume set LNCS 15046 and 15047 constitutes the refereed proceedings of the 17th International Symposium, ISVC 2024, held at Lake Tahoe, NV, USA, during October 21-23, 2024.
The 54 full papers and 12 poster papers were carefully reviewed and selected from 120 submissions. A total of 8 papers were also accepted for oral presentation in special tracks from 15 submissions. The papers cover the following topical sections:
Part I: Deep Learning; Computer Graphics; Video Analysis and Event Recognition; Motion and Tracking; Detection and Recognition; Visualization, and Medical Image Analysis.
Part II: Segmentation; Recognition; Generalization in Visual Machine Learning; Vision and Robotics for Agriculture; Virtual Reality; Applications, and Poster.

Inhalt

Deep Learning I: Advanced Post-Processing for Object Detection Dataset Generation.- AFIDAF: Alternating Fourier and Image Domain Adaptive Filters as an Effcient Alternative to Attention in ViTs.- Multi-Actor-Critic Deep Reinforcement Learning with Hindsight Experience Replay.- Improving Zero-Shot Template-Based 6D Pose Estimation with Geometric Features.- Contrastive Loss based on Contextual Similarity for Image Classification. Computer Graphics: Estimation of Global Illumination using Cycle-Consistent Adversarial Networks.- Anisotropic Point Synthesis by Example.- 3D Fluid Shape Control by Direct Manipulation.- An Epithelium-Inspired Deformation Modeling Framework for 4D Sheets.- Hybrid Voxel Formats for Effcient Ray Tracing. Video Analysis and Event Recognition: PIEPredict++: An Improved Pedestrian Intention Estimation Model Incorporating Comprehensive Environment Information.- Infant Video Interaction Recognition using Monocular Depth Estimation.- Real-Time Predictor in Two-Players Fighting Game via Vision Transformer.- Explainable Action-Recognition based Approach for Unsupervised Video Anomaly Detection.- LORTSAR: Low-Rank Transformer for Skeleton-based Action Recognition. Motion and Tracking: Pedestrian tracking using ankle-level 2D-LiDAR based on ByteTrack.- Human Pose Estimation-Based ID Assignment Method in the Wild: A Real-Time Approach.- RGB-T-UV Multi-Modal Object Tracking Based on Transformer Network.- Evaluating the Impact of Dehazing Algorithms on Visual Object Tracking Performance.- MEM: Mask Enhancement Model for Video Object Segmentation. Detection and Recognition: Analysis Automation for High Explosive Breakout Symmetry.- CLAP: Concave Linear APproximation for Quadratic Graph Matching.- Generalizing Neural Radiance Fields for Robust 6D Pose Estimation of Unseen Appearances.- Effectiveness of Residual Noise based Methods for Single Image based Morphing Attack Detection: A Comparative Study.- Black Box Adversarial Face Transformation Network. Deep Learning II: Bi-Feature Selection Deep Learning-based Techniques for Speech Emotion Recognition.- ActiveConfusion: A Time-Effcient Approach to the Cold-Start Problem in Active Learning by Incorporating Confusion from Pretext Task.- Removing Adverse Volumetric Effects from Trained Neural Radiance Fields.- Thermal Image Synthesis: Bridging the Gap between Visible and Infrared Spectrum.- Anomaly Detection in Mutual Actions: Unsupervised Classification of Fighting and Non-Fighting Behaviors using Transformer-based Variational Autoencoder.- Visualization: Visualizing Polarization Effects of Gravitational Waves Using Particle Rings and Surfaces in Virtual Reality.- Seeing is Believing: The Role of Scatterplots in Recommender System Trust and Decision-Making.- Interactive Visual Analysis of Camouflaged Objects.- GAIA: A Benchmark of Analyzing User Rankings for Synthesized Images. Medical Image Analysis: Motion and Light Artifact Mitigation for Remote PPG with Noise-Aware Post-Processor Network.- J-Net: A Low-Resolution Lightweight Neural Network for Semantic Segmentation in the Medical field for Embedded Deployment.- Accurate Remote PPG Waveform Recovery from Video Using a Multi-Task Learning Temporal Model.- Effcient Lung Segmentation for Tumour Detection.