AI Paper Atlas

Research map / Image generation / Image restoration

RFD-ECNet: Extreme Underwater Image Compression with Reference to Feature Dictionary

Mengyao Li, Liquan Shen, Peng Ye, Guorui Feng, Zheyin Wang

ICCV 2023 · Image generation › Image restoration · approach: Inverse imaging with learned priors · 9 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 USNB style normalization of dictionary features (ablation)RFD-ECNet · UGWI testBD-rate (%): -33.6 -> -53.7Improved
add reference feature variant module (RFVM) (ablation)RFD-ECNet · UGWI testBD-rate (%): -43.1 -> -53.7Improved
add both USNB and RFVM (ablation)RFD-ECNet · UGWI testBD-rate (%): -24.6 -> -53.7Improved
match dictionary at scales 2,3,4 instead of scale 2 only (ablation)RFD-ECNet · UGWI testBD-rate (%): -46.1 -> -53.7Improved
reference a feature dictionary instead of using video codec inter modeUWI compression · UGWI test (Minnen anchor)BD-rate (%): -22.1 -> -59.5Improved
use inter mode instead of intra mode in video codecsVVC and HEVC · UGWI test (Minnen anchor)BD-rate (%): -21.0 -> -22.1 (VVC)No clear effect
narrow channels from 192 to 128 (slim version)RFD-ECNet · Average of four UWI testsetsBD-rate (%): -48.5 -> -43.0Mixed

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