Discriminative unimodal feature selection and fusion for RGB-D salient object detection

•We propose a novel RGB-D salient object detection model, which takes the equalities of input into consideration.•An SG-MWMG sub-network is designed to determine the informative and non-informative regions in the input RGB-D images.•An MCFF module is designed to fuse the unimodal RGB and depth features at multiple scales and levels.