This volume constitutes the revised selected papers of 31st International Workshop on Frontiers of Computer Vision, IW-FCV 2025, in Seoul, South Korea, during February 1012, 2025.

Inhalt

Extracting Temporal Features Robust to Headwear Variation from Video Sequences of Body Sway for Person Identification.- Multi-group Vision Semantic Centroid for Semantic Segmentation.- A Cross-Modal Knowledge Distillation Approach for RGB-to-Infrared Video Action Recognition.- Automating Wood Grain Sensory Evaluation via Image Feature Fusion Approach using Vision-Language Models.- Behavior Analysis of Grokking with Flooding.- A Pseudo Optical-Axis-Matching Projector-Camera System by Using Intermediate Viewpoint Image Generation.- Multi-View Images Acquisition for Digital Content Creation of Cultural Heritage.- Cherry Packing Skill Proficiency Classification Using Hand Motion Analysis.- Visual Feedback Control of an Underactuated Hand for Grasping Objects of Varying Stiffness.- CATS+: Cultural heritage data Augmentation and Transformation System with Correlation Analysis.- Mask-Guided Region-Specific Editing in Diffusion Models.- Lighting Condition Discrimination Based on Hyperspectral Images.- Reflectance and Illumination Estimation from A Known Reflectance for Classification.- Detection of Red Tide Plankton in Automatic Microscope System Imaging Using Deep Learning.- Self-Attention-Driven Refinements for Training-Free Segmentation in Diffusion Transformers.- Character Restoration System for Shihai-monjo in Ancient Document Images Based on Image Inpainting Technique.- From Single to Unified Training: Evaluating Multi-Organ Segmentation Techniques.- Occlusion Robust 2D Pose-Based Gait Recognition.- Vision-based Human Fall Detection with Pre-trained YOLO Models: Comparison and Application.- Improving Impurity Detection in Laver Images through Data Augmentation.- 3D Annotation System for Work Assistance in Living Environments.- Personal Protective Equipment Detection Based on Improved YOLO11-Nano using Multi-Perspective Attention.- Preliminary experiments for behavior log generation considering spatial relationship between actions and objects.