AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

LookupViT: Compressing visual information to a limited number of tokens

Rajat Koner, Gagan Jain, Sujoy Paul, Volker Tresp, Prateek Jain

ECCV 2024 · Architectures and efficiency › Attention and transformers · approach: Hybrid and interpretable ViT designs · 14 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 bidirectional information flow (MHBC compressed-to-lookup) to LookupViT (ablation)LookupViT B/16 5x5 · ImageNet-1k (from scratch)top-1 accuracy: 69.7 -> 79.1Improved
add lookup tokens with cross-attention to compressed tokens (ablation)LookupViT B/16 5x5 · ImageNet-1k (from scratch)top-1 accuracy: 69.2 -> 79.1Improved
add classification loss on lookup tokens (ablation)LookupViT B/16 5x5 · ImageNet-1k (from scratch)top-1 accuracy: 78.2 -> 79.1Improved
add classification loss on compressed tokens (ablation)LookupViT B/16 5x5 · ImageNet-1k (from scratch)top-1 accuracy: 78.4 -> 79.1Improved
build compressed tokens by parameter-free resize of lookup tokens instead of random learnable tokens (ablation)LookupViT B/16 5x5 · ImageNet-1k (from scratch)top-1 accuracy: 77.9 -> 79.1Improved
replace ViT with LookupViT 5x5 (B/16, from scratch)ViT B/16 · ImageNet-1k (from scratch)top-1 accuracy and GFLOPs: 78.6 -> 79.1 accuracy, 35.1 -> 16.5 GFLOPsImproved
replace ViT with LookupViT 3x3 after IN-21k pretrainingViT B/16 · ImageNet-1k (IN-21k pretrain, finetune)top-1 accuracy: 83.7 -> 77.9Hurt
use LookupViT instead of ViT on corrupted/OOD imagesViT B/16 · ImageNet-Caccuracy: 56.4 -> 59.9Improved
use LookupViT 5x5 instead of ViT on ImageNet-AViT B/16 · ImageNet-Aaccuracy: 6.7 -> 5.4Hurt
use LookupViViT instead of ViViT on SSv2 at half FLOPsViViT-B · Something-Something V2 (1 view)accuracy: 51.2 -> 56.3Improved
use LookupViViT instead of ViViT on Kinetics400ViViT-B · Kinetics400 (12 views)accuracy: 78.6 -> 77.9No clear effect
train with random compressed-token resolution (Multi-Res)LookupViT · ImageNet-1k (IN-21k pretrain, finetune)top-1 accuracyImproved

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