Full Description
This two-volume set constitutes the proceedings of the 20th International Symposium, ISVC 2025, held in Las Vegas, NV, USA, during November 17-19, 2025.
The 54 full papers and 18 poster papers were carefully reviewed and selected from 118 submissions. The papers cover the following topical sections:
Part I: Deep Learning; Computer Graphics; Motion and Tracking; Applications; Object Detection and Recognition; Medical Imaging, and Virtual Reality.
Part II: Segmentation; 3D; Recognition; Video Analysis and Event Recognition; Biometrics; Visualization, and Poster.
Contents
.- Deep Learning.
.- Effective Fine-Tuning with Eigenvector Centrality Based Pruning.
.- Continual Learning for Image Captioning through Improved Image-Text Alignment.
.- NeuroShield-ViT: Mechanistic Understandings of Representation Vulnerabilities and Engineering Robust Vision Transformers.
.- CATCH: A Modular Cross-domain Adaptive Template with Hook.
.- PPP: Performance Preserving Pruning for Deep Neural Networks.
.- Computer Graphics.
.- Joint Neural SDF Reconstruction and Semantic Segmentation for CAD Models.
.- Scheduling the Off-Diagonal Weingarten Loss of Neural SDFs for CAD Models.
.- PPREDICTING UNCANNY PERCEPTION IN VIRTUAL HUMANS FACES THROUGH IMAGE FEATURES.
.- Extreme Views: 3DGS Filter for Novel View Synthesis from Out-of-Distribution Camera Poses.
.- Indoor Heat Estimation from a Single Visible-Light Panorama.
.- Motion and Tracking.
.- Long-Term Prediction of Local and Global Human Motion with Occlusion Recovery.
.- Enhanced Vehicle Speed Detection Considering Lane Recognition Using Drone Videos in California.
.- A Video-Based Fish Counting System for Juvenile Fish Transferred with Water Flow in Aquaculture.
.- A Geometry-Driven Approach to Detecting and Tracking Multiple Aerial Objects.
.- Synthetic Data Enabled Under the Cover Human Pose Estimation using RGB Images.
.- Applications.
.- Object-Based Camera Pose Estimation from a Single Object Detection and Gravity Vector.
.- Leveraging Multimodal LLMs for Building Condition Assessment from Street-View Imagery.
.- Taming Inversion Drift: A Dynamic Look-ahead Controller for Rectified Flow Models.
.- Variable Rate Image Compression via N-Gram Context based Swin-transformer.
.- Abnormality Detection in Maize Fields Using Selective Domain Adaptation-Driven Data Augmentation.
.- Object Detection and Recognition.
.- UniRare: Framework Reconciling Bottom-Up and Top-Down Attention in Visual Saliency.
.- The Hidden Cost of Poor Annotations: How Label Quality Affects Camouflaged Object Detection Performance.
.- Self-supervised structured object representation learning.
.- YOLOv8-Based Detection and Classification of Plant Cells in Light Micrographs.
.- Task-Conditioned Next-Fixation Prediction in Assembly Tasks.
.- Medical Imaging.
.- ColonMeshNet: Fusing Local and Global Context for Accurate 3D Colon Segmentation.
.- Leveraging Autoencoder-Based Filters to Enhance Automated Medical Classifiers.
.- Interpretable Tile-Based Classification of Paclitaxel Exposure.
.- Self-Supervised Representation Learning for AI-Based Musculoskeletal Radiograph Registry Construction.
.- Features for Measuring the Progression of Gastric Atrophy Focused on Gastric Areae Shadow Pattern in X-ray Images of Stomach.
.- Virtual Reality.
.- Gaze Prediction in Virtual Reality Without Eye Tracking Using Visual and Head Motion Cues.
.- Examining Visual Discomfort Associated with Perceived Depth Changes in Optical See-Through Augmented Reality.
.- Evaluating Stepwise Increases in Degrees of Freedom from Walking to Manipulation for Immersive Analytics.
.- Shadow-Based Touch Detection for Virtual Forearm Keyboards in AR Text Entry.
.- LiFMCR: Dataset and Benchmark for Light Field Multi-Camera Registration.
