The two-volume set 15725 and 15726 constitutes the refereed proceedings of the 23rd Scandinavian Conference on Image Analysis, SCIA 2025, held in Reykjavik, Iceland, during June 2325, 2025.
The 60 revised full papers presented in these proceedings were carefully reviewed and selected from 98 submissions. The papers are organized in the following topical sections:

Part I: Machine learning and deep learning; Segmentation, Grouping, and Shape; Vision for robotics and autonomous vehicles; Biometrics, faces, body gestures and pose; 3D vision from multiview and other sensors; and Vision applications and systems.
Part II: Datasets and evaluation; Image and video processing, analysis, and understanding; Detection, recognition, classification, and localization in 2D and/or 3D; Medical, biological, and cell microscopy; Explainable AI for CV; Vision + language (+ other modalities); Computational photography; Fairness; and Low·level and physics·based vision.

Inhalt

.- Datasets and evaluation
.- FIORD: A Fisheye Indoor-Outdoor Dataset with LIDAR Ground Truth for 3D Scene Reconstruction and Benchmarking.
.- SunVid: A Curated Online Video Database for Sundown Syndrome Research .
.- Comparative Analysis of rPPG and Motion-Based Approaches for Heart and Respiration Rate Estimation from Videos.
.- Aligning Subjective and Objective Assessments in Super-Resolution Models.
.- Comparing Next-Day Wildfire Predictability of MODIS and VIIRS Satellite Data.
.- Image and video processing, analysis, and understanding.
.- A spectral-preserving zero-shot technique for hyperspectral pansharpening.
.- Characteristic-driven Deep Learning in Synthetic Aperture Radar Target Recognition.
.- Single-Image Localised Reflection Removal with k-Order Differences Term.
.- Detection, recognition, classification, and localization in 2D and/or 3D.
.- The Impact of Semi-Supervised Learning on Line Segment Detection.
.- Weak Cube R-CNN: Weakly Supervised 3D Detection using only 2D Bounding Boxes.
.- Training Binary Neural Networks with Deep Semantic Guidance.
.- Revisiting Likelihood-Based Out-of-Distribution Detection by Modeling Representations.
.- Quantifying Epistemic Uncertainty in Absolute Pose Regression.
.- EfficientPose 6D: Scalable and Efficient 6D Object Pose Estimation.
.- Subspace-Based Embedded Feature Reduction for Fast Anomaly Detection.
.- Medical, biological, and cell microscopy.
.- Active Learning with nnUNet for Coronary Artery Lumen Segmentation using a Centerline Prior.
.- Automated Cardiac Adipose Tissue Segmentation in Computed Tomography: A Literature Review.
.- Determining Fetal Orientations From Blind Sweep Ultrasound Video.
.- A Comparison of Deep Learning Methods for Cell Detection in Digital Cytology.
.- Assessing the Efficacy of Multi-Task Learning in Mammographic Density Classification: A study on Class Imbalance and Model Performance.
.- Fine-Grained Classification of Unpigmented Skin Cancer from Paired Dermatoscopy Images.
.- Explainable AI for CV.
.- A Meaningful Perturbation Metric for Evaluating Explainability Methods.
.- The Weighting Game: Evaluating Quality of Explainability Methods..
.- Vision + language (+ other modalities).
.- POEM: Precise Object-level Editing via MLLM control.
.- Food State Recognition from Recipes Using Multimodal Model for Task Monitoring in Autonomous Cooking Robots.
.- Computational photography.
.- Data-Efficient Limited-Angle Tomography Using Deep Priors and Regularization.
.- Fairness.
.- Are generative models fair? A study of racial bias in dermatological image generation.
.- Low·level and physics·based vision.
.- Demosaicing and Neural Network Based Image Reconstruction in the Presence of Noise.