Can I read Graph-based Vision Transformer for Small Datasets on EtoBox?

Graph-based Vision Transformer for Small Datasets by yangkunkuo is a document available to read on EtoBox.

What is Graph-based Vision Transformer for Small Datasets about?

The document presents the Graph-based Vision Transformer (GvT), which addresses the performance gap between Vision Transformers (ViTs) and Convolutional Neural Networks (CNNs) when trained on small datasets. GvT employs graph convolutional projection and talking-heads technology to enhance local feature representation and reduce low-rank bottlenecks in attention mechanisms. Experimental results indicate that GvT achieves competitive or superior performance compared to CNNs and traditional ViTs without requi

Author

yangkunkuo

Language

EN