DNA: Deeply-supervised Nonlinear Aggregation for Salient Object Detection
This work addresses a bottleneck in salient object detection for computer vision applications, representing an incremental improvement over existing methods.
The paper tackles the suboptimal linear aggregation of multi-scale features in salient object detection by proposing Deeply-supervised Nonlinear Aggregation (DNA), which improves performance by aggregating features nonlinearly, achieving favorable results against state-of-the-art methods on various datasets.
Recent progress on salient object detection mainly aims at exploiting how to effectively integrate multi-scale convolutional features in convolutional neural networks (CNNs). Many popular methods impose deep supervision to perform side-output predictions that are linearly aggregated for final saliency prediction. In this paper, we theoretically and experimentally demonstrate that linear aggregation of side-output predictions is suboptimal, and it only makes limited use of the side-output information obtained by deep supervision. To solve this problem, we propose Deeply-supervised Nonlinear Aggregation (DNA) for better leveraging the complementary information of various side-outputs. Compared with existing methods, it i) aggregates side-output features rather than predictions, and ii) adopts nonlinear instead of linear transformations. Experiments demonstrate that DNA can successfully break through the bottleneck of current linear approaches. Specifically, the proposed saliency detector, a modified U-Net architecture with DNA, performs favorably against state-of-the-art methods on various datasets and evaluation metrics without bells and whistles.