Learning Correlation Structures for Vision Transformers

• URL: http://arxiv.org/abs/2404.03924v1
• Date: Fri, 5 Apr 2024 07:13:28 GMT
• Title: Learning Correlation Structures for Vision Transformers
• Authors: Manjin Kim, Paul Hongsuck Seo, Cordelia Schmid, Minsu Cho,
• Abstract summary: We introduce a new attention mechanism, dubbed structural self-attention (StructSA) We generate attention maps by recognizing space-time structures of key-query correlations via convolution. This effectively leverages rich structural patterns in images and videos such as scene layouts, object motion, and inter-object relations.
• Score: 93.22434535223587
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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.

Related papers

• SIM-Trans: Structure Information Modeling Transformer for Fine-grained Visual Categorization [59.732036564862796]
  We propose the Structure Information Modeling Transformer (SIM-Trans) to incorporate object structure information into transformer for enhancing discriminative representation learning. The proposed two modules are light-weighted and can be plugged into any transformer network and trained end-to-end easily. Experiments and analyses demonstrate that the proposed SIM-Trans achieves state-of-the-art performance on fine-grained visual categorization benchmarks.
  arXiv  Detail & Related papers  (2022-08-31T03:00:07Z)
• Structured Attention Composition for Temporal Action Localization [99.66510088698051]
  We study temporal action localization from the perspective of multi-modality feature learning. Unlike conventional attention, the proposed module would not infer frame attention and modality attention independently. The proposed structured attention composition module can be deployed as a plug-and-play module into existing action localization frameworks.
  arXiv  Detail & Related papers  (2022-05-20T04:32:09Z)
• Vision Transformer with Convolutions Architecture Search [72.70461709267497]
  We propose an architecture search method-Vision Transformer with Convolutions Architecture Search (VTCAS) The high-performance backbone network searched by VTCAS introduces the desirable features of convolutional neural networks into the Transformer architecture. It enhances the robustness of the neural network for object recognition, especially in the low illumination indoor scene.
  arXiv  Detail & Related papers  (2022-03-20T02:59:51Z)
• Structure-Regularized Attention for Deformable Object Representation [17.120035855774344]
  Capturing contextual dependencies has proven useful to improve the representational power of deep neural networks. Recent approaches that focus on modeling global context, such as self-attention and non-local operation, achieve this goal by enabling unconstrained pairwise interactions between elements. We consider learning representations for deformable objects which can benefit from context exploitation by modeling the structural dependencies that the data intrinsically possesses.
  arXiv  Detail & Related papers  (2021-06-12T03:10:17Z)
• Adaptive Interaction Modeling via Graph Operations Search [109.45125932109454]
  We automate the process of structures design to learn adaptive structures for interaction modeling. We experimentally demonstrate that our architecture search framework learns to construct adaptive interaction modeling structures. Our method achieves competitive performance with state-of-the-arts.
  arXiv  Detail & Related papers  (2020-05-05T13:01:09Z)
• Look-into-Object: Self-supervised Structure Modeling for Object Recognition [71.68524003173219]
  We propose to "look into object" (explicitly yet intrinsically model the object structure) through incorporating self-supervisions. We show the recognition backbone can be substantially enhanced for more robust representation learning. Our approach achieves large performance gain on a number of benchmarks, including generic object recognition (ImageNet) and fine-grained object recognition tasks (CUB, Cars, Aircraft)
  arXiv  Detail & Related papers  (2020-03-31T12:22:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.