RGB-D Saliency Detection via Cascaded Mutual Information Minimization
This work addresses the challenge of improving saliency detection accuracy in computer vision applications, though it is incremental as it builds on existing multi-modal methods with a novel regularization approach.
The paper tackles the problem of ineffective multi-modal learning in RGB-D saliency detection by introducing a cascaded framework using mutual information minimization to reduce redundancy between RGB and depth features, achieving state-of-the-art results on benchmark datasets and contributing a new large dataset with multiple annotation types.
Existing RGB-D saliency detection models do not explicitly encourage RGB and depth to achieve effective multi-modal learning. In this paper, we introduce a novel multi-stage cascaded learning framework via mutual information minimization to "explicitly" model the multi-modal information between RGB image and depth data. Specifically, we first map the feature of each mode to a lower dimensional feature vector, and adopt mutual information minimization as a regularizer to reduce the redundancy between appearance features from RGB and geometric features from depth. We then perform multi-stage cascaded learning to impose the mutual information minimization constraint at every stage of the network. Extensive experiments on benchmark RGB-D saliency datasets illustrate the effectiveness of our framework. Further, to prosper the development of this field, we contribute the largest (7x larger than NJU2K) dataset, which contains 15,625 image pairs with high quality polygon-/scribble-/object-/instance-/rank-level annotations. Based on these rich labels, we additionally construct four new benchmarks with strong baselines and observe some interesting phenomena, which can motivate future model design. Source code and dataset are available at "https://github.com/JingZhang617/cascaded_rgbd_sod".