AI Paper Atlas

Research map / Image generation / Editing and style transfer

Geometrically Constrained Stenosis Editing in Coronary Angiography via Entropic Optimal Transport

Jialin Li, Zhuo Zhang, Cao Yue, Guipeng Lan, Jiabao Wen, Shuai Xiao, Jiachen Yang

ICML 2026 · Image generation › Editing and style transfer · approach: Inversion-based diffusion editing · 0 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
edit in composite vessel-structure domain instead of edge domain (ablation)OT-Bridge Editor · Edited regionDice: 0.592 -> 0.892Improved
edit in composite domain instead of segmentation domain (ablation)OT-Bridge Editor · Edited regionDice: 0.767 -> 0.892Improved
add non-edited region preservation constraint (ablation)OT-Bridge Editor · Outside maskSSIM: 0.786 -> 0.878Improved
add non-edited region preservation constraint (ablation)OT-Bridge Editor · Outside maskLPIPS: 0.317 -> 0.248Improved
apply geometric constraints along the whole generation path (GPG) instead of endpoint only (ablation)OT-Bridge Editor · Stenosis editingboundary Dice: 0.765 -> 0.895Improved
add the ∂m term to GPG (ablation)OT-Bridge Editor · Stenosis editingboundary Dice: 0.582 -> 0.895Improved
GPG vs endpoint on final trajectory error (ablation)OT-Bridge Editor · Stenosis editingfinal trajectory error E_T: 2.8 -> 1.1Improved
use OT-Bridge Editor instead of SiameseDiffSiameseDiff · CAG stenosis editingmIoU: 0.837 -> 0.892Improved
add synthetic data to real training for YOLOv8YOLOv8 · ARCADEmAP@0.5: 0.525 -> 0.727Improved
add synthetic data to real training for Grounding DINOGrounding DINO · ARCADEmAP@0.5: 0.276 -> 0.418Improved
add synthetic data to real training on multi-center dataYOLOv8 · Multi-centermAP@0.5: 0.654 -> 0.731Improved
train on synthetic only vs real onlyYOLOv8 · Multi-centermAP@0.5: 0.654 -> 0.582Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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