The four-volume set LNCS 16068-16071 constitutes the proceedings of the 34th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2025, held in Kaunas, Lithuania, September 9-12, 2025. The 170 full papers and 8 abstracts included in these conference proceedings were carefully reviewed and selected from 375 submissions. The conference strongly values the synergy between theoretical progress and impactful real-world applications, and actively encourages contributions that demonstrate how artificial neural networks are being used to address pressing societal and technological challenges.

Table of Contents:
.- HARNet: Human Activity Recognition with Spatial-temporal Features. .- Wavelet-based Self-image Blending for More General Face Forgery Detection. .- HCNQA: Enhancing 3D Visual Question-Answering with Hierarchical Concentration Narrowing Supervision. .- Text-Image Encoder-based Contrastive Regression for AI-Generated Image Quality Assessment. .- Deblurring with Improved Video Diffusion Model. .- Understanding Image Classification Prediction with Any Segment Explanation. .- F2Unet: F-shaped U-Net Architecture for Medical Image Segmentation Combining Fourier Transforms. .- Efficient Real-Time On-Mobile Video Super-Resolution with Automatic Evolutionary Neural Architecture Search. .- Swin-DAG-VNet for Fetal Head Segmentation and Elliptical Parameter Regression for Circumference Measurement. .- GPIS: Geometric Informed Polygon Prompt for Incision Segmentation. .- Topological Enhancement Learning Module for Segmentation of Complex and Irregular Structures in 3D Medical Imaging. .- Multi-Representation Adapter with Neural Architecture Search forEfficient Range-Doppler Radar Object Detection. .- LPSF-LiDARNet: Log-Polar Spatiotemporal Fusion-Based LiDAR Point Cloud Semantic Segmentation for Autonomous Driving. .- PSRDET: Fast Multimodal Detecton based on Prior Scene Repair for All-weather Road Sensing. .- Art Style Backdoor Attacks on Semantic Segmentation Models. .- CLIP-Guided Frequency-Aware Representation Learning for Generalizable Remote-Sensing Image Tampering Detection. .- Generalized Object Detection in the Infrared Domain based on Common-Representation Learning. .- Balance Discriminability and Integrality for Robust Salient Object Detection. .- Unsupervised Domain Adaptive Hand Mesh Reconstruction of 2D Images in the Wild. .- Towards Context-Aware Compositional Zero-Shot Food Recognition via SalientFusion. .- SparWR: A Lightweight Architecture for Medical Grayscale Image Super-Resolution. .- PaRCE: Probabilistic and Reconstruction-based Competency Estimation for CNN-based Image Classification. .- Sparse Attention Diffusion Model for Pathological Micrograph Deblurring. .- RDIF: Infrared and Visible Image Fusion Based on Reverse Cross-Attention and Diffusion Model. .- RDGE-6D: Reverse Direction Geometry Injection for 6D Pose Estimation. .- YOLO11n BSS: Examination Room Behaviors Detection Based on Improved YOLO11n Model. .- Efficient Long-Term Motion Feature Learning via Frequency-Based Key Frame Guidance for Action Recognition. .- Region Expansion: Optimization of Patch-fetching Method for Point Cloud Denoising. .- SAMTNU: Adaptive Segment Anything Model for Thyroid and Nodule Ultrasound Image Segmentation. .- AgileIR: Memory-Efficient Group Shifted Windows Attention for Lightweight Image Restoration. .- Constrained Learnable Channel-wise Normalization for Single-Source Domain Generalization in Medical Image Segmentation. .- FaceSnap: Enhanced ID-fidelity Network for Tuning-free Portrait Customization. .- Improving the Transferability of Point Cloud Attack via Spectral-aware Admix and Optimization Designs. .- IHCP:Image Hiding against Blind Compression Based on Quality Prediction. .- VideoPCDNet: Video Parsing and Prediction with Phase Correlation Networks. .- Uncertainty Quantification in Video Distortion Classification under Dataset Shift. .- PIMSeg: Point Cloud Segmentation via 2D Image Mapping and Multimodal Feature Integration. .- Prototype-Guided Local Spatial Attention for Model Explanation. .- Robustness Verification for Object Detectors using Set-Based Reachability Analysis. .- Incorporating Feature Pyramid Tokenization and Open Vocabulary Semantic Segmentation. .- PaFi-GS: Gaussian Splatting via Propagation-Aware Filtering for Urban Street View Rendering. .- Vision-Text Interaction with Orientation-Awareness for Referring Remote Sensing Image Segmentation. .- U-FQA: A Unified Face Quality Assessment Score for Improved Unknown Identity Detection in Facial Recognition Systems. .- DFU-Net: A Diffusion-based Fourier Neural Operator-aided U-Net Model for Medical Image Segmentation in Edge Devices. .- An Enhanced Dual-Stream Architecture for State-of-the-Art Artist and Style Classification. .- A Scene Text Detection Method Based on Supervised Contrastive Learning. .- CM-MNet: A Coordinate Space-Aware Mamba-Based Multi-Task Model for 3D Fine Lesions in Elongated Structures Segmentation and Diagnosis in MS and NMOSD. .- Problem-Driven and Shape-Guided: Multi-scale Deform KAN for X-shaped Anterior Visual Pathway Segmentation. .- A Camouflaged Object Detection Network with Global Cross-space Perception and Flexible Local Feature Refinement Network. .- See Beyond: Benchmarking MLLMs’ Visual Relational Reasoning Ability. .- HeteroCap: Hierarchical Visual-Semantic Fusion with Heterogeneous Graphs for Image Captioning. .- Dual-Head Feature Enhancement for Graph-Based Cross-View Multi-Object Tracking. .- Semantic Enhanced Interaction for Unsupervised Cross-modal Hashing Retrieval.