PreprintCC BY-NC-ND 4.0arXiv
Background2Background: a self-supervised background removal framework for high dimensional microscopy
Abstract
High-dimensional microscopy captures features across space, time, and spectrum, but complex backgrounds obscure signals. We present Background2Background (B2B), a self-supervised background removal framework leveraging the backgrounds larger correlation relative to the signal. Local pixel shuffling disrupts signal correlations while preserving background structure, transforming signals into effective noise for Noise2Noise-based background removal. Across fluorescence, stimulated Raman scattering and mid-infrared photothermal imaging, B2B achieves over 40-fold background suppression and reveals hidden signatures.
Related
- Learning Transferable Self-Supervised Priors for Super-Resolution Reconstruction in Structured Illumination Microscopy
- Open quantum battery in the background of a three-dimensional rotating black hole
- A unified self-supervised framework for single-frame Fresnel CDI and overlapped ptychography
- Three-dimensional visualization of threading dislocation in GaN by polarized-light microscopy
- Scattering-force-driven light-sheet fluorescence microscopy for scan-free three-dimensional imaging
BibTeX
@misc{ding2026background2background,
title = {{Background2Background: a self-supervised background removal framework for high dimensional microscopy}},
author = {Guangrui Ding and Haonan Lin and Rong Tang and Bethany Weinberg and Jianpeng Ao and Qing Xia and Guorong Hu and Dingcheng Sun and Jiaze Yin and Lei Tian and Ji-Xin Cheng},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.17615},
eprint = {2610.17615},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.17615},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - Background2Background: a self-supervised background removal framework for high dimensional microscopy
AU - Guangrui Ding
AU - Haonan Lin
AU - Rong Tang
AU - Bethany Weinberg
AU - Jianpeng Ao
AU - Qing Xia
AU - Guorong Hu
AU - Dingcheng Sun
AU - Jiaze Yin
AU - Lei Tian
AU - Ji-Xin Cheng
PY - 2026
DA - 2026/10/13
PB - arXiv
DO - 10.48550/arxiv.2610.17615
UR - https://arxiv.org/abs/2610.17615
AB - High-dimensional microscopy captures features across space, time, and spectrum, but complex backgrounds obscure signals. We present Background2Background (B2B), a self-supervised background removal framework leveraging the backgrounds larger correlation relative to the signal. Local pixel shuffling disrupts signal correlations while preserving background structure, transforming signals into effective noise for Noise2Noise-based background removal. Across fluorescence, stimulated Raman scattering and mid-infrared photothermal imaging, B2B achieves over 40-fold background suppression and reveals hidden signatures.
ER -