Visual and Textual Deep Feature Fusion for Document Image Classification

Open paper

Benchmark	Model	Rank	Results
document-image-classification-on-rvl-cdip	Cross-Modal	–	Accuracy: 97.05%Parameters: 197M