AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Contextformer: A Transformer with Spatio-Channel Attention for Context Modeling in Learned Image Compression

A. Burakhan Koyuncu, Han Gao, Atanas Boev, Georgii Gaikov, Elena Alshina, Eckehard Steinbach

ECCV 2022 · Architectures and efficiency › Attention and transformers · approach: Attention head and block redesign

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 cross-channel attention by increasing channel segments N_cs (ablation)Contextformer · Kodakrate savingsImproved
use channel-first order instead of spatial-first order (ablation)Contextformer · KodakBD-rateImproved
use spatial-first order at low bitrate (bpp<0.3) (ablation)Contextformer · KodakBD-rateMixed
use more than four channel segments (ablation)Contextformer · KodakBD-rateNo clear effect
increase network depth beyond 8 layers (ablation)Contextformer · KodakBD-rateNo clear effect
increase depth L and embedding size d_e together (ablation)Contextformer · KodakBD-rateImproved
use BDS instead of DS in encoder (ablation)Contextformer · Kodak encoding timetime (s): 56 -> 32Improved
use BDS and SCS instead of DS in encoder (ablation)Contextformer · 4K encoding timetime (s): 1240 -> 120Improved
use wavefront coding instead of DS in decoder (ablation)Contextformer · Kodak decoding timetime (s): 62 -> 44Improved
use wavefront coding instead of DS in decoder (ablation)Contextformer · 4K decoding timetime (s): 1440 -> 820Improved
replace Cui et al. context model with Contextformer at similar sizeCui et al. model · KodakBD-rate: 10.1%Improved
replace Minnen&Singh with ContextformerMinnen&Singh · KodakBD-rate vs VTM 16.2: 3.7% -> 6.9%Improved

Later papers that compared with this method

1 later accepted paper reports Contextformer as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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