Paper Image

RGB-D salient object detection using transformers and CNNs

Published on:

3 July 2023

Primary Category:

Computer Vision and Pattern Recognition

Paper Authors:

Kang Yi,

Jing Xu,

Xiao Jin,

Fu Guo,

Yan-Feng Wu

Bullets

Key Details

Proposes a dual-stream asymmetric framework with transformers for RGB and CNNs for depth

Models high-order representations to distinguish RGB and depth

Designs spatial and channel fusion modules for different stages

Employs a cascaded decoder to integrate multi-scale features

AI generated summary

RGB-D salient object detection using transformers and CNNs

This paper proposes a new approach for detecting salient objects in RGB-D images, using transformer and convolutional neural network (CNN) architectures. It introduces a dual-stream asymmetric framework to handle the different characteristics of RGB and depth data. The method achieves strong performance by modeling high-order representations and using specialized fusion techniques.

Answers from this paper

Comments

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

Sign up to comment on this paper

Sign Up