AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Edit2Perceive: Image Editing Diffusion Models Are Strong Dense Perceivers

Yiqing Shi, Yiren Song, Mike Zheng Shou

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 3 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 image-editing base (FLUX.1 Kontext) instead of text-to-image FLUX.1 (ablation)Edit2Perceive depth, basic config · NYUv2AbsRel: 6.8 -> 5.1Improved
use image-editing base instead of T2I base with full recipe (ablation)Edit2Perceive depth · KITTIAbsRel: 8.4 -> 7.9Improved
use image-editing base instead of T2I for normals (ablation)Edit2Perceive normals · NYUv2 normalsmean angular error: 16.4 -> 15.7Improved
add pixel-space consistency loss (ablation)T2I FLUX.1 depth · NYUv2AbsRel: 6.8 -> 5.4Improved
add pixel-space consistency loss (ablation)FLUX.1 Kontext depth · NYUv2AbsRel: 4.7 -> 4.4Improved
add pixel-space consistency loss (ablation)FLUX.1 Kontext depth, uniform mapping · KITTIAbsRel: 9.6 -> 9.6No clear effect
add pixel-space consistency loss (ablation)FLUX.1 Kontext normals · NYUv2 normals11.25 deg accuracy: 60.0 -> 61.6Improved
use sqrt depth mapping instead of uniform (ablation)FLUX.1 Kontext depth, no LCons · KITTIAbsRel: 9.6 -> 8.2Improved
use sqrt depth mapping instead of uniform (ablation)T2I FLUX.1 depth, no LCons · KITTIAbsRel: 13.2 -> 10.2Improved
increase inference steps beyond 4 (ablation)Edit2Perceive · depth and normal benchmarksdelta1 / 11.25 deg accuracyHurt
single-step inference instead of multi-step (ablation)Edit2Perceive · depth and normal benchmarksaccuracyNo clear effect

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