Bifurcated backbone strategy for RGB-D salient object detection

Abstract

Multi-level feature fusion is a fundamental topic in computer vision. It has been exploited to detect, segment and classify objects at various scales. When multi-level features meet multi-modal cues, the optimal feature aggregation and multi-modal learning strategy become a hot potato. In this paper, we leverage the inherent multi-modal and multi-level nature of RGB-D salient object detection to devise a novel cascaded refinement network. In particular, first, we propose to regroup the multi-level features into teacher and student features using a bifurcated backbone strategy (BBS). Second, we introduce a depth-enhanced module (DEM) to excavate informative depth cues from the channel and spatial views. Then, RGB and depth modalities are fused in a complementary way. Our architecture, named Bifurcated Backbone Strategy Network (BBS-Net), is simple, efficient, and backbone-independent. Extensive experiments show that BBS-Net significantly outperforms eighteen SOTA models on eight challenging datasets under five evaluation measures, demonstrating the superiority of our approach (∼4% improvement in S-measure vs. the top-ranked model: DMRA-iccv2019). In addition, we provide a comprehensive analysis on the generalization ability of different RGB-D datasets and provide a powerful training set for future research.

Depth Quality-Inspired Feature Manipulation for Efficient RGB-D Salient Object Detection

“…With the popularity of depth sensors/devices, RGB-D SOD has become a hot research topic [18,20,39,53,55,60], because additional useful spatial information embedded in depth maps could serve as a complementary cue for more robust detection [64]. Meanwhile, depth data is already widely available on many mobile devices [17], e.g., modern smartphones like Huawei Mate 40 Pro, iPhone 12 Pro, Samsung Galaxy S20+.…”