paper-with-me

Papers

Dynamic Group Transformer: A General Vision Transformer Backbone with Dynamic Group Attention

2022-03-08 · Kai Liu, Tianyi Wu, Cong Liu, Guodong Guo

Recently, Transformers have shown promising performance in various vision tasks. To reduce the quadratic computation complexity caused by each query attending to all keys/values, various methods have constrained the range of attention within local regions, where each query only attends to keys/values within a hand-crafted window. However, these hand-crafted window partition mechanisms are data-agnostic and ignore their input content, so it is likely that one query maybe attends to irrelevant keys/values. To address this issue, we propose a Dynamic Group Attention (DG-Attention), which dynamically divides all queries into multiple groups and selects the most relevant keys/values for each group. Our DG-Attention can flexibly model more relevant dependencies without any spatial constraint that is used in hand-crafted window based attention. Built on the DG-Attention, we develop a general vision transformer backbone named Dynamic Group Transformer (DGT). Extensive experiments show that our models can outperform the state-of-the-art methods on multiple common vision tasks, including image classification, semantic segmentation, object detection, and instance segmentation.

📄 PDF Abstract BibTeX arXiv:2203.03937

Code (0)

등록된 구현이 없습니다.

Tasks

image-classificationImage ClassificationInstance Segmentationobject-detectionObject DetectionSemantic Segmentation

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Adam 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

Similar Papers 제목 키워드 기반

Towards Lightweight Transformer via Group-wise Transformation for Vision-and-Language Tasks

2022-04-16 · Gen Luo, Yiyi Zhou, Xiaoshuai Sun, Yan Wang 외

Despite the exciting performance, Transformer is criticized for its excessive parameters and computation cost. However, compressing Transformer remains as an open problem due to its internal complexity of the layer desig…

image-classificationImage Classification

Self-attention in Vision Transformers Performs Perceptual Grouping, Not Attention

2023-03-02 · Paria Mehrani, John K. Tsotsos

Recently, a considerable number of studies in computer vision involves deep neural architectures called vision transformers. Visual processing in these models incorporates computational models that are claimed to impleme…

Saliency Detection

GPViT: A High Resolution Non-Hierarchical Vision Transformer with Group Propagation

2022-12-13 · Chenhongyi Yang, Jiarui Xu, Shalini De Mello, Elliot J. Crowley 외

We present the Group Propagation Vision Transformer (GPViT): a novel nonhierarchical (i.e. non-pyramidal) transformer model designed for general visual recognition with high-resolution features. High-resolution features …

Decoderimage-classificationImage ClassificationInstance Segmentation+5

A Unified Framework for Vision Transformers Equivariant to Discrete Subgroups of $\mathrm{O}(2)$

2026-06-26 · Tīkun Ông, Georg Bökman arxiv

Vision transformers have become a dominant architecture for visual recognition. However, standard models do not explicitly encode the planar symmetries that arise in many vision domains. We introduce a family of vision t…

Platonic Transformers: A Solid Choice For Equivariance

2025-10-03 · Mohammad Mohaiminul Islam, Rishabh Anand, David R. Wessels, Friso de Kruiff 외 arxiv

While widespread, Transformers lack inductive biases for geometric symmetries common in science and computer vision. Existing equivariant methods often sacrifice the efficiency and flexibility that make Transformers so e…

Molecular Property PredictionPoint Clouds