Can I read HIRI-ViT: Scaling Vision Transformer with High Resolution Inputs on EtoBox?

HIRI-ViT: Scaling Vision Transformer with High Resolution Inputs by Yao, Ting; Li, Yehao; Pan, Yingwei; Mei, Tao is a scholarly article available to read on EtoBox.

What is HIRI-ViT: Scaling Vision Transformer with High Resolution Inputs about?

The hybrid deep models of Vision Transformer (ViT) and Convolution Neural Network (CNN) have emerged as a powerful class of backbones for vision tasks. Scaling up the input resolution of such hybrid backbones naturally strengthes model capacity, but inevitably suffers from heavy computational cost that scales quadratically. Instead, we present a new hybrid backbone with HIgh-Resolution Inputs (namely HIRI-ViT), that upgrades prevalent four-stage ViT to five-stage ViT tailored for high-resolution inputs. HIRI-ViT is built upon the seminal idea of decomposing the typical CNN operations into two parallel CNN branches in a cost-efficient manner. One high-resolution branch directly takes primary high-resolution features as inputs, but uses less convolution operations. The other low-resolution branch first performs down-sampling and then utilizes more convolution operations over such low-resolution features. Experiments on both recognition task (ImageNet-1K dataset) and dense prediction tasks (COCO and ADE20K datasets) demonstrate the superiority of HIRI-ViT. More remarkably, under comparable computational cost ($\sim$5.0 GFLOPs), HIRI-ViT achieves to-date the best published Top-1 accura

Author

Yao, Ting; Li, Yehao; Pan, Yingwei; Mei, Tao

Published

2024

Language

EN