About this document

CAFNET: Infrared & Low-Light Image Fusion by okuwobi is a document available to read on EtoBox.

This paper proposes a novel cross-attention fusion network (CAFNET) to fuse infrared and low illumination visible-light images. CAFNET extracts features from the images using the first layer of a pre-trained VGG16 network. It then calculates cross-attention between the features to distinguish differences and spatial attention to reflect image characteristics. Weight maps are generated from the attentions to pre-fuse the images in the network. Finally, detail injection with Gaussian blur is applied to furthe

Author

okuwobi

Language

EN