Siamese Network for RGB-D Salient Object Detection and Beyond

26 Aug 2020  ·  Keren Fu, Deng-Ping Fan, Ge-Peng Ji, Qijun Zhao, Jianbing Shen, Ce Zhu ·

Existing RGB-D salient object detection (SOD) models usually treat RGB and depth as independent information and design separate networks for feature extraction from each. Such schemes can easily be constrained by a limited amount of training data or over-reliance on an elaborately designed training process... Inspired by the observation that RGB and depth modalities actually present certain commonality in distinguishing salient objects, a novel joint learning and densely cooperative fusion (JL-DCF) architecture is designed to learn from both RGB and depth inputs through a shared network backbone, known as the Siamese architecture. In this paper, we propose two effective components: joint learning (JL), and densely cooperative fusion (DCF). The JL module provides robust saliency feature learning by exploiting cross-modal commonality via a Siamese network, while the DCF module is introduced for complementary feature discovery. Comprehensive experiments using five popular metrics show that the designed framework yields a robust RGB-D saliency detector with good generalization. As a result, JL-DCF significantly advances the state-of-the-art models by an average of ~2.0% (max F-measure) across seven challenging datasets. In addition, we show that JL-DCF is readily applicable to other related multi-modal detection tasks, including RGB-T (thermal infrared) SOD and video SOD, achieving comparable or even better performance against state-of-the-art methods. We also link JL-DCF to the RGB-D semantic segmentation field, showing its capability of outperforming several semantic segmentation models on the task of RGB-D SOD. These facts further confirm that the proposed framework could offer a potential solution for various applications and provide more insight into the cross-modal complementarity task. read more

PDF Abstract

Results from the Paper


Ranked #2 on RGB-D Salient Object Detection on SIP (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
RGB-D Salient Object Detection DES JL-DCF* S-Measure 93.6 # 4
Average MAE 0.021 # 5
max E-Measure 97.5 # 2
max F-Measure 92.9 # 2
RGB-D Salient Object Detection NJU2K JL-DCF* S-Measure 91.1 # 4
Average MAE 0.040 # 8
max E-Measure 94.8 # 3
max F-Measure 91.3 # 4
RGB-D Salient Object Detection NLPR JL-DCF* S-Measure 92.6 # 4
Average MAE 0.023 # 3
max F-Measure 91.7 # 4
max E-Measure 96.4 # 2
RGB-D Salient Object Detection SIP JL-DCF* S-Measure 89.2 # 2
max E-Measure 94.9 # 1
max F-Measure 90.0 # 2
Average MAE 0.046 # 4
RGB-D Salient Object Detection STERE JL-DCF* S-Measure 91.1 # 2
Average MAE 0.039 # 4
max F-Measure 90.7 # 3
max E-Measure 94.9 # 1

Methods


No methods listed for this paper. Add relevant methods here