FVeinSyn is a large‑scale synthetic finger‑vein image generator that separates vascular topology synthesis from imaging appearance rendering. It uses stochastic L‑systems to create anatomically valid, identity‑distinctive vein patterns, a cascaded region‑aware GAN to produce realistic near‑infrared images, and an intra‑class diversity generator to simulate realistic variations. The framework generated 500,000 images across 10,000 identities, and models trained with this data outperformed real‑data‑only baselines on eight public datasets, improving average accuracy by 27.43%.
Vein recognition is a secure biometric technology often constrained by limited annotated data and imaging variations. While data augmentation mitigates this, strategies designed for natural images may disrupt the fine-grained topology and textures essential for identity discrimination.
OpenVeinNet is a finger vein verification framework that tackles open-set scenarios by combining Dynamic Snake Convolution, which extracts local curvilinear vein structures through adaptive sampling, with a graph convolutional backbone that models long-range topological relationships between vein regions. The model introduces a Centroid Angular Hybrid Loss to promote intra-class compactness and inter-class angular separation in the embedding space. Experiments on five public datasets under leave-one-dataset-out training demonstrate strong cross-dataset generalisation, low equal error rates, and competitive true accept rates at fixed false accept rates.
Finger vein verification is a promising biometric modality for secure authentication because vascular patterns are internal, difficult to observe externally, and relatively resistant to presentation a...
arXiv:2607. 23371v1 Announce Type: cross Abstract: Vascular segmentation is a standard procedure for clinical diagnosis, yet the specific visual features determining model decisions remain poorly understood.
Vascular segmentation is a standard procedure for clinical diagnosis, yet the specific visual features determining model decisions remain poorly understood. This paper investigates the visual cues Convolutional Neural Networks (CNNs) use to segment blood vessels across two distinct imaging domains: fluorescence microscopy and retinal fundus photography.
arXiv:2607. 22139v1 Announce Type: cross Abstract: Accurate pixel-level classification of coronary angiograms is critical for cardiovascular disease assessment, yet the field lacks standardized evaluation protocols.
arXiv:2603. 18846v3 Announce Type: replace-cross Abstract: Foundation models are used to extract transferable representations from large amounts of unlabeled data, typically via self-supervised learning (SSL).
ReG-SAM is a SAM-based framework designed for 2D vessel segmentation in medical images. It introduces reference graph prompt embeddings (GPEs) and vascular prototype embeddings (VPEs) to capture global spatial and fine-grained modality-specific vessel features, respectively. By building a modality-wise vascular database and learning these embeddings from reference masks, ReG-SAM consistently outperforms existing baselines across 19 datasets, especially on thin vessels.
The paper introduces StyleGANCA, a lightweight neural cellular automata (NCA) based generative adversarial network designed for medical image synthesis. By combining a StyleGAN-inspired mapping network with adaptive style modulation in a multi-scale NCA framework, the model achieves high-quality image generation with far fewer parameters than existing adversarial, variational, diffusion, and NCA baselines. Experiments on BloodMNIST and PathMNIST show competitive FID and KID scores, and the synthetic images preserve class-specific information, effectively supporting downstream multi-class classifier training.
Accurate dermatological diagnosis naturally necessitates equitable performance across diverse populations, yet a systematic lack of expertly annotated images, especially for underrepresented skin tones and rare diseases, impedes progress toward measurably fair methods. We introduce cgDDI (Controllable Generation of Diverse Dermatological Imagery), a hybrid framework that (1) synthesizes realistic healthy skin samples without disturbing other input properties, (2) maps single-sample rare lesions onto novel skin-tones and locations non-parametrically, and (3) allows for efficient parametric generation with as few as 10 training samples.
arXiv:2609.01598v1 Announce Type: new Abstract: Accurate segmentation of vascular structures in digital subtraction angiography (DSA) images remains challenging due to the thin, elongated, and branch...
arXiv:2606. 15837v1 Announce Type: cross Abstract: Deep neural networks (DNNs) frequently fail to generalize to out-of-distribution (OOD) medical images because of variations in scanners and acquisition protocols.