Papers › Learning Correlation Structures for Vision Transformers

Learning Correlation Structures for Vision Transformers

5 Apr 2024CVPR 2024 1arXiv:2404.03924archive 2025-07-28

Manjin Kim, Paul Hongsuck Seo, Cordelia Schmid, Minsu Cho

We introduce a new attention mechanism, dubbed structural self-attention (StructSA), that leverages rich correlation patterns naturally emerging in key-query interactions of attention. StructSA generates attention maps by recognizing space-time structures of key-query correlations via convolution and uses them to dynamically aggregate local contexts of value features. This effectively leverages rich structural patterns in images and videos such as scene layouts, object motion, and inter-object relations. Using StructSA as a main building block, we develop the structural vision transformer (StructViT) and evaluate its effectiveness on both image and video classification tasks, achieving state-of-the-art results on ImageNet-1K, Kinetics-400, Something-Something V1 & V2, Diving-48, and FineGym.

PaperPDFConference PDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Action ClassificationAction RecognitionObjectVideo Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Action Classification Kinetics-400 StructViT-B-4-1 Acc@1 83.4 #68 of 207 Archive leaderboard report
Action Recognition Diving-48 StructVit-B-4-1 Accuracy 88.3 #6 of 18 Archive leaderboard report
Action Recognition Something-Something V1 StructVit-B-4-1 Top 1 Accuracy 61.3 #8 of 74 Archive leaderboard report
Action Recognition Something-Something V2 StructVit-B-4-1 Top-1 Accuracy 71.5 #28 of 123 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AttentionConvolutionDense ConnectionsLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmaxVision Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections