47 repos across 2 sub-areas
Deep learning models and tools for extracting interpretable features from pathology images, particularly whole-slide images (WSIs) for histopathology analysis. The cluster centers on vision transformer and CNN-based architectures optimized for medical image understanding, with libraries like timm providing backbone implementations. Repositories range from foundational models like CONCH and GigaPath (which appear as central nodes) to application-specific feature extraction pipelines for computational pathology tasks.
Cluster 648219
33 repos
Vision Transformer Models & Feature Extraction
14 repos
PyTorch-based vision transformer architectures and pre-trained models for image understanding and feature extraction. This cluster centers on modern computer vision models like DINOv2 and Vision Transformers (ViT), along with tools for working with model weights (safetensors format) and extracting learned image representations. Repositories here span both foundational transformer models and applications built on top of them, useful for anyone working on image classification, representation learning, or transfer learning in vision tasks.