Research map / Image generation / Image restoration
λSplit: Self-Supervised Content-Aware Spectral Unmixing for Fluorescence Microscopy
Federico Carrara, Mehdi Seifi, Florian Jug
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace pixel-wise RLU unmixing with λSplit | RLU · CellAtlas noise, 2ms exposure (32 bands) | PSNR: 23.98 -> 27.14 | Improved |
| replace pixel-wise RLU unmixing with λSplit | RLU · CellAtlas noise, 5ms exposure (32 bands) | PSNR: 28.91 -> 34.03 | Improved |
| replace pixel-wise RLU unmixing with λSplit | RLU · CellAtlas noise, 20ms exposure (32 bands) | PSNR: 36.68 -> 36.46 | No clear effect |
| replace pixel-wise RLU unmixing with λSplit | RLU · CellAtlas noise, 20ms exposure (32 bands) | LPIPS: 0.125 -> 0.155 | Hurt |
| replace pixel-wise RLU unmixing with λSplit | RLU · CellAtlas, 3 bands (same per-band SNR) | PSNR: 26.39 -> 27.25 | Improved |
| replace pixel-wise RLU unmixing with λSplit | RLU · CellAtlas, 32 bands (same total photon budget) | PSNR: 26.15 -> 31.32 | Improved |
| replace NMF-RI with λSplit | NMF-RI · CellAtlas, 3 bands (same per-band SNR) | LPIPS: 0.237 -> 0.283 | Hurt |
| reduce spectral bands from 32 to 3 at constant per-band SNR | λSplit · CellAtlas (same per-band SNR) | PSNR: 36.46 -> 27.25 | Hurt |
| reduce spectral bands from 32 to 3 at constant per-band SNR | LU · CellAtlas (same per-band SNR) | PSNR: 35.91 -> 23.69 | Hurt |
| add sum-to-one constraint to non-negative linear unmixing (FCLU) | NNLU · CellAtlas noise, 5ms exposure (32 bands) | PSNR: 28.55 -> 22.03 | Hurt |
Closest papers by meaning
- scSplit: Bringing Severity Cognizance to Image Decomposition in Fluorescence Microscopy (NeurIPS 2025 · similarity 0.90)
- Unmixing Diffusion for Self-Supervised Hyperspectral Image Denoising (CVPR 2024 · similarity 0.86)
- FluoCLIP: Stain-Aware Focus Quality Assessment in Fluorescence Microscopy (CVPR 2026 · similarity 0.85)
- V2V3D: View-to-View Denoised 3D Reconstruction for Light Field Microscopy (CVPR 2025 · similarity 0.84)
- SR-CACO-2: A Dataset for Confocal Fluorescence Microscopy Image Super-Resolution (NeurIPS 2024 · similarity 0.84)
- SheafStain: Sheaf-Theoretic Schr\"odinger Bridge for Spatially and Biologically Coherent Virtual Staining (NeurIPS 2026 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
