AI Paper Atlas

Research map / Image generation / Image restoration

Perceiving and Modeling Density for Image Dehazing

Tian Ye, Yunchen Zhang, Mingchao Jiang, Liang Chen, Yun Liu, Sixiang Chen, Erkang Chen

ECCV 2022 · Image generation › Image restoration · approach: Dehazing and deraining · 161 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
add haze density encoding matrix (vs identity matrix) (ablation)PMDNet · Haze4KPSNR: 30.33 -> 33.49Improved
use unsupervised density encoding matrix instead of GT transmission map (ablation)PMDNet · Haze4KPSNR: 31.42 -> 33.49Improved
use unsupervised density encoding matrix instead of estimated transmission map (ablation)PMDNet · Haze4KPSNR: 31.17 -> 33.49Improved
use density encoding matrix instead of normalized difference map (ablation)PMDNet · Haze4KPSNR: 32.21 -> 33.49Improved
replace residual blocks with SHA attention (ablation)Baseline dehazing net · Haze4KPSNR: 24.40 -> 26.39Improved
use SHA instead of SWRCA attention (ablation)Baseline dehazing net · Haze4KPSNR: 25.87 -> 26.39Improved
use SHA instead of FA attention (ablation)Baseline dehazing net · Haze4KPSNR: 25.30 -> 26.39Improved
use FA attention instead of CBA (ablation)Baseline dehazing net · Haze4KPSNR: 25.36 -> 25.30Hurt
extend SHA to multi-branch MHAB (ablation)Baseline dehazing net · Haze4KPSNR: 26.39 -> 27.02Improved
add contextual transformer branch (MHA-C vs MHAB) (ablation)Baseline dehazing net · Haze4KPSNR: 27.02 -> 27.58Improved
PMDNet vs second-best methodPMDNet · Haze4KPSNR: 28.53 -> 33.49Improved

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