AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Restoring Images in Adverse Weather Conditions via Histogram Transformer

Shangquan Sun, Wenqi Ren, Xinwei Gao, Rui Wang, Xiaochun Cao

ECCV 2024 · Architectures and efficiency › Attention and transformers · approach: Efficient image transformers · 162 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 dynamic-range convolution (pixel sorting before convolution) instead of vanilla convolution (ablation)Histoformer · Outdoor-RainPSNR: 31.94 -> 32.08Improved
change sorting order of dynamic-range convolution (vertical-horizontal versus horizontal-vertical) (ablation)Histoformer · Outdoor-RainPSNR: 32.03 vs 32.08No clear effect
use DHSA instead of top-k sparse attention (TKSA) (ablation)Histoformer · Outdoor-RainPSNR: +0.96 dBImproved
use both BHR and FHR branches in DHSA instead of a single branchDHSA · Outdoor-RainPSNRImproved
use DGFF instead of mixed-scale feed-forward network (MSFN) (ablation)Histoformer · Outdoor-RainPSNR: +0.3 dBImproved
add Pearson correlation lossHistoformer · Outdoor-RainPSNRImproved
vary the weight of the correlation lossHistoformer · Outdoor-RainPSNRNo clear effect
increase the number of bins and channels (C x B from 12 to 44)DHSA · Outdoor-RainPSNRImproved

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