AI Paper Atlas

Research map / Image generation / Image restoration

Taming Generative Diffusion Model for Task-Oriented Infrared Imaging

Tengyu Ma, Zhilong Dai, Yubo Diao, Guanming An, Long Ma, Jinyuan Liu, Risheng Liu

CVPR 2026 · Image generation › Image restoration · approach: Pansharpening and spectral fusion

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
add spectral regularization to base loss (no semantic loss) (ablation)single-step infrared diffusion · HM-TIR restorationLPIPS: 0.315 -> 0.224Improved
add semantic alignment loss to base loss (ablation)single-step infrared diffusion · FMB segmentationmIoU: 0.403 -> 0.442Improved
add spectral loss on top of base and semantic losses (ablation)single-step infrared diffusion · HM-TIR restorationLPIPS: 0.372 -> 0.1857Improved
add semantic alignment loss to spectral variant (ablation)single-step infrared diffusion · HM-TIR restorationPSNR: 25.703 -> 25.684No clear effect
add semantic alignment loss to spectral variant (ablation)single-step infrared diffusion · FMB segmentationmIoU: 0.417 -> 0.447Improved
add learnable task prompt p_tau (ablation)task-aware LoRA · M3FD detectionmAP: 0.682 -> 0.718Improved
use dynamic timestep estimation instead of fixed timesteps (50, 250, 500, 1000) (ablation)single-step infrared diffusion · HM-TIR restorationPSNRImproved
prompt-only tuning versus full-model fine-tuning on unseen scenesDiffBIR comparison · TNO and RoadSceneVIFImproved
reformulate as single-step diffusiondiffusion restoration · computational costFLOPs (T): 32.577 -> 1.191 vs SUPIRImproved
propose method versus PPFN on SSIMsingle-step infrared diffusion · HM-TIR Composited NormalSSIM: 0.8160 (PPFN) vs 0.8021Hurt

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