AI Paper Atlas

Research map / Image generation / Image restoration

Teaching Tailored to Talent: Adverse Weather Restoration via Prompt Pool and Depth-Anything Constraint

Sixiang Chen, Tian Ye, Kai Zhang, Zhaohu Xing, Yunlong Lin, Lei Zhu

ECCV 2024 · Image generation › Image restoration · approach: Dehazing and deraining · 22 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 prompt pool for weather prompts (ablation)T3-DiffWeather · Outdoor-RainPSNR: 31.05 -> 31.99Improved
use prompt pool instead of shared learnable prompts (ablation)T3-DiffWeather · Outdoor-RainPSNR: 31.38 -> 31.99Improved
add matched keys for sub-prompt selection (ablation)prompt pool · Outdoor-RainPSNR: 31.72 -> 31.99Improved
add general prompts for scene modeling (ablation)T3-DiffWeather · Outdoor-RainPSNR: 31.52 -> 31.99Improved
constrain general prompts with Depth-Anything features (ablation)general prompts · Outdoor-RainPSNR: 31.67 -> 31.99Improved
use Depth-Anything instead of DINOv2 (ablation)general prompts · Outdoor-RainPSNR: 31.82 -> 31.99Improved
add contrastive prompt loss (ablation)T3-DiffWeather · Outdoor-RainPSNR: 31.71 -> 31.99Improved
scale Depth-Anything from ViT-S to ViT-L (ablation)general prompt constraint · Outdoor-RainPSNR: 31.99 -> 32.06No clear effect
increase general prompt length from 256 to 320 (ablation)general prompts · Outdoor-RainPSNR: 31.99 -> 32.03No clear effect
increase sub-prompt length from 64 to 128 (ablation)prompt pool · Outdoor-RainPSNR: 31.99 -> 32.04No clear effect
reconstruct degradation residual with 2 sampling steps instead of WeatherDiffusion 25 stepsdiffusion restoration · 256x256 complexityGFLOPs: 248.4G x 25 steps -> 59.82G x 2 stepsImproved

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