MicroCLIP: Connecting Text and Microscopy Images via Contrastive Language--Image Pretraining
cs.CV
Submitted: 2026-10-13
Updated: 2026-10-13
Code: https://github.com/Y2R-mv1p/MicroCLIP
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Cytoarchitecture in Words: Weakly Supervised Vision-Language Modeling for Human Brain Microscopy
- SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models