Deeply-Supervised Recurrent Convolutional Neural Network for Saliency Detection

TL;DR

This paper introduces a deeply-supervised recurrent convolutional neural network (DSRCNN) for saliency detection, effectively integrating local, global, and contextual information to produce high-quality saliency maps.

Contribution

The novel DSRCNN combines recurrent connections and deep supervision within a VGGNet-16 based architecture for improved saliency detection.

Findings

Outperforms state-of-the-art methods on five benchmark datasets.

Effectively captures local, global, and contextual information.

Produces more accurate and robust saliency maps.

Abstract

This paper proposes a novel saliency detection method by developing a deeply-supervised recurrent convolutional neural network (DSRCNN), which performs a full image-to-image saliency prediction. For saliency detection, the local, global, and contextual information of salient objects is important to obtain a high quality salient map. To achieve this goal, the DSRCNN is designed based on VGGNet-16. Firstly, the recurrent connections are incorporated into each convolutional layer, which can make the model more powerful for learning the contextual information. Secondly, side-output layers are added to conduct the deeply-supervised operation, which can make the model learn more discriminative and robust features by effecting the intermediate layers. Finally, all of the side-outputs are fused to integrate the local and global information to get the final saliency detection results. Therefore, the DSRCNN combines the advantages of recurrent convolutional neural networks and deeply-supervised nets. The DSRCNN model is tested on five benchmark datasets, and experimental results demonstrate that the proposed method significantly outperforms the state-of-the-art saliency detection approaches on all test datasets.