TNT

Transformer in Transformer

Computer VisionIntroduced 200012 papers
Source Paper

Description

Transformer is a type of self-attention-based neural networks originally applied for NLP tasks. Recently, pure transformer-based models are proposed to solve computer vision problems. These visual transformers usually view an image as a sequence of patches while they ignore the intrinsic structure information inside each patch. In this paper, we propose a novel Transformer-iN-Transformer (TNT) model for modeling both patch-level and pixel-level representation. In each TNT block, an outer transformer block is utilized to process patch embeddings, and an inner transformer block extracts local features from pixel embeddings. The pixel-level feature is projected to the space of patch embedding by a linear transformation layer and then added into the patch. By stacking the TNT blocks, we build the TNT model for image recognition.

Image source: Han et al.

Papers Using This Method

Quadratic Gaussian Splatting for Efficient and Detailed Surface Reconstruction2024-11-25Intensity Field Decomposition for Tissue-Guided Neural Tomography2024-11-01Towards Minimal Targeted Updates of Language Models with Targeted Negative Training2024-06-19Swin transformers are robust to distribution and concept drift in endoscopy-based longitudinal rectal cancer assessment2024-05-06Revolutionizing Traffic Sign Recognition: Unveiling the Potential of Vision Transformers2024-04-29Nested-TNT: Hierarchical Vision Transformers with Multi-Scale Feature Processing2024-04-20A Mini-Block Fisher Method for Deep Neural Networks2022-02-08PyramidTNT: Improved Transformer-in-Transformer Baselines with Pyramid Architecture2022-01-04TnT Attacks! Universal Naturalistic Adversarial Patches Against Deep Neural Network Systems2021-11-19Tree in Tree: from Decision Trees to Decision Graphs2021-10-01Tensor Normal Training for Deep Learning Models2021-06-05Transformer in Transformer2021-02-27