Paper Image

Dual Path Rain Removal Network

Published on:

16 January 2024

Primary Category:

Computer Vision and Pattern Recognition

Paper Authors:

Bingcai Wei


Key Details

Proposes a two-branch network - CNN and vision transformer

Includes an attention fusion module to selectively combine features

Ablation studies validate the dual-path design

Comparisons show superiority over other rain removal methods

AI generated summary

Dual Path Rain Removal Network

This paper proposes a dual-path neural network for removing rain from images. It has a convolutional neural network (CNN) branch to model local features and a vision transformer branch to model long-range dependencies. An attention-based fusion module selectively combines features from both branches. Through ablation studies and comparisons to other methods, the dual-path design is shown to be more effective than single-branch approaches.

Answers from this paper


No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up