Patch Transformer

Patch Transformers represent a burgeoning area of research that adapts the Transformer architecture to process data divided into smaller patches, improving efficiency and enabling the capture of both local and global features. Current research focuses on optimizing patch sizes and strategies for various data types, including time series, images, and 3D point clouds, with models like Medformer and MultiResFormer demonstrating advancements in specific applications. This approach offers significant potential for improving performance and reducing computational costs in diverse fields, ranging from medical image analysis and time series forecasting to hyperspectral image processing and semantic segmentation.

Papers

March 19, 2022

HIPA: Hierarchical Patch Transformer for Single Image Super Resolution
Qing Cai, Yiming Qian, Jinxing Li, Jun Lv, Yee-Hong Yang, Feng Wu, David Zhang
Vision Transformer Single Image Super Resolution Transformer Based Architecture Proton Beam Patch Transformer Novel Transformer Architecture

March 13, 2022

Sparse Local Patch Transformer for Robust Face Alignment and Landmarks Inherent Relation Learning
Jiahao Xia, Weiwei qu, Wenjian Huang, Jianguo Zhang, Xi Wang, Min Xu
Facial Landmark Relation Pattern Face Alignment Patch Transformer Relation Learning

February 27, 2022

TransKD: Transformer Knowledge Distillation for Efficient Semantic Segmentation
Ruiping Liu, Kailun Yang, Alina Roitberg, Jiaming Zhang, Kunyu Peng, Huayao Liu, Yaonan Wang, Rainer Stiefelhagen
Knowledge Distillation Semantic Segmentation Benchmark Patch Transformer Efficient Semantic Segmentation Large Pre Trained Transformer

Patch Transformer

Papers

HIPA: Hierarchical Patch Transformer for Single Image Super Resolution

Sparse Local Patch Transformer for Robust Face Alignment and Landmarks Inherent Relation Learning

TransKD: Transformer Knowledge Distillation for Efficient Semantic Segmentation