AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

MaxViT: Multi-axis Vision Transformer

Zhengzhong Tu, Hossein Talebi, Han Zhang, Feng Yang, Peyman Milanfar, Alan Bovik, Yinxiao Li

ECCV 2022 · Architectures and efficiency › Attention and transformers · approach: Efficient and robust vision transformers · 1,240 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 grid-attention in stage 1 (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.36 -> 83.62Improved
add grid-attention in stage 2 (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.38 -> 83.62Improved
add grid-attention in stage 3 (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.00 -> 83.62Improved
use grid-attention rather than block-attention in stage 1 (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.49 -> 83.62Improved
use grid-attention rather than block-attention in stage 3 (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.40 -> 83.62Improved
add MBConv in stage 1 (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.24 -> 83.62Improved
add MBConv in stage 3 (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 82.65 -> 83.62Improved
add MBConv in stage 4 (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.09 -> 83.62Improved
place MBConv before block- and grid-attention (C-BA-GA) (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.07 -> 83.62Improved
order block-attention before grid-attention (C-BA-GA) instead of C-GA-BA (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 83.54 -> 83.62No clear effect
use sequential instead of parallel multi-axis attention (ablation)MaxViT-T · ImageNet-1Ktop-1 accuracy: 82.64 -> 83.62Improved
use sequential instead of parallel multi-axis attention (ablation)MaxViT-L · ImageNet-1Ktop-1 accuracy: 83.54 -> 85.17Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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