A Saliency Enhanced Feature Fusion Based Multiscale RGB-D Salient Object Detection Network

Multiscale convolutional neural network (CNN) has demonstrated remarkable capabilities in solving various vision problems. However, fusing features of different scales always results in large model sizes, impeding the application of multiscale CNNs in RGB-D saliency detection. In this paper, we propose a customized feature fusion module, called Saliency Enhanced Feature Fusion (SEFF), for RGB-D saliency detection. SEFF utilizes saliency maps of the neighboring scales to enhance the necessary features for fusing, resulting in more representative fused features. Our multiscale RGB-D saliency detector uses SEFF and processes images with three different scales. SEFF is used to fuse the features of RGB and depth images, as well as the features of decoders at different scales. Extensive experiments on five benchmark datasets have demonstrated the superiority of our method over ten SOTA saliency detectors.

1. INTRODUCTION

RGB-D Salient object detection (SOD) can use depth information to extract salient objects with similar colors to background in RGB images, which has wide applications in computer vision problems, such as object tracking [1], [2], image retrieval [3], [4], and instance segmentation [5], [6].