AI Paper Atlas

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

Steered Diffusion: A Generalized Framework for Plug-and-Play Conditional Image Synthesis

Nithin Gopalakrishnan Nair, Anoop Cherian, Suhas Lohit, Ye Wang, Toshiaki Koike-Akino, Vishal M. Patel, Tim K. Marks

ICCV 2023 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 12 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
steer implicit clean-image prediction x_0|t instead of noisy x_tSteered Diffusion · CelebA face super-resolutionFID: 62.24 -> 51.19Improved
steer implicit prediction on semantic generation versus ILVRSteered Diffusion · CelebA semantic layout to faceFID: 107.46 -> 41.38Improved
Steered Diffusion versus trained methods on semantic mIoUSteered Diffusion · CelebA semantic layout to facemIoU: 0.723 vs 0.532Hurt
use multi-step implicit modulation (iterate N times per timestep)Steered Diffusion · inpainting and colorizationconditioning qualityImproved
use time-varying scaling factor k(t) instead of fixedSteered Diffusion · semantic generationsample realismImproved
use very high or very low scaling factor KSteered Diffusion · semantic generationsample qualityHurt
use a trained noise-robust inverse mapper instead of implicit denoised predictioncolorization · face colorizationdetail preservationHurt
Steered Diffusion versus ILVR for colorizationSteered Diffusion · face colorizationFID: 67.66 -> 49.72Improved
Steered Diffusion versus PULSE on NIQESteered Diffusion · CelebA face super-resolutionNIQE: 5.04 vs 9.24Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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