AI Paper Atlas

Research map / Image generation / Image restoration

Computationally-Efficient Neural Image Compression with Shallow Decoders

Yibo Yang, Stephan Mandt

ICCV 2023 · Image generation › Image restoration · approach: Learned image compression · 44 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 SGA iterative encoding to the two-layer shallow synthesis (ablation)2-layer synthesis · KodakBD rate savings vs BPG (%): -5.19 -> 4.67Improved
use two-layer nonlinear synthesis instead of JPEG-like linear synthesis (ablation)shallow decoder · KodakBD rate savings vs BPG (%): -20.95 -> -5.19Improved
increase JPEG-like synthesis kernel size to k=26 (ablation)JPEG-like synthesis · Kodakbit-rate vs linear CNN synthesis: within 1% of linear CNN with 94% fewer FLOPsImproved
use minimal kernel size k=s=16 in JPEG-like synthesis (ablation)JPEG-like synthesis · Kodakvisual artifactsHurt
replace ELIC analysis transform with Mean-Scale Hyperprior analysis (ablation)2-layer synthesis · Kodakbit-rate vs BPG: over 6% worseHurt
add residual connection to the two-layer synthesis (ablation)2-layer synthesis · KodakBD rate savingsImproved
use inverse GDN activation and larger output kernel size (ablation)2-layer synthesis · KodakBD rate savingsImproved
evaluate shallow synthesis on perceptual metrics MS-SSIM and LPIPS2-layer synthesis + SGA · Kodak, Tecnick, CLICBD-rate vs Hyperprior: 8% to 12% worseHurt
pair a shallow decoder with a stronger encoder and iterative encodingMean-Scale Hyperprior · Kodakdecoding complexity: 108.97 -> 20.52 KMAC per pixelImproved

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