AI Paper Atlas

Research map / Detection and segmentation / CNN features

Tiling artifacts and trade-offs of feature normalization in the segmentation of large biological images

Elena Buglakova, Anwai Archit, Edoardo D'Imprima, Julia Mahamid, Constantin Pape, Anna Kreshuk

ICCV 2025 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 10 citations (OpenAlex)

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 changedWhereMeasuredEffect
use BatchRenorm instead of InstanceNorm (ablation)3D U-Net · organoids matrixDice (eval mode): 0.61 -> 0.85Improved
use BatchRenorm instead of InstanceNorm (ablation)3D U-Net · embryo foregroundDice (eval mode): 0.67 -> 0.74Improved
use BatchRenorm instead of InstanceNorm (ablation)3D U-Net · organoids boundariesDice (eval mode): 0.75 -> 0.73Hurt
use BatchRenorm instead of BatchNorm (ablation)3D U-Net · organoids boundariesDice (eval mode): 0.49 -> 0.73Improved
use BatchRenorm instead of BatchNorm (ablation)3D U-Net · organoidstrain/eval disparity: 0.48 -> 0Improved
use BatchRenorm instead of InstanceNorm (ablation)3D U-Net · organoidstile mismatch: 0.11 -> noImproved
use BatchRenorm instead of InstanceNorm (ablation)3D U-Net · embryotile mismatch: 0.16 -> noImproved
use BatchRenorm instead of InstanceNorm (ablation)UNETR · organoids matrixDice (eval mode): 0.10 -> 0.87Improved
use BatchRenorm instead of BatchNorm (ablation)UNETR · organoids matrixDice (eval mode): 0.03 -> 0.87Improved
use BatchRenorm instead of InstanceNorm (ablation)3D U-Net transfer embryo->plants · plants foregroundDice (eval mode): 0.50 -> 0.50No clear effect
use BatchRenorm instead of InstanceNorm (ablation)3D U-Net transfer plants->embryo · embryo foregroundDice (eval mode): 0.54 -> 0.59Improved
remove feature normalization (Identity) instead of BatchRenorm (ablation)3D U-Net · plants foregroundDice (eval mode): 0.89 -> 0.52Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy