Boundary-Guided Feature Aggregation Network for Salient Object Detection

Abstract

Fully convolutional networks (FCN) has significantly improved the performance of many pixel-labeling tasks, such as semantic segmentation and depth estimation. However, it still remains non-trivial to thoroughly utilize the multi-level convolutional feature maps and boundary information for salient object detection. In this paper, we propose a novel FCN framework to integrate multi-level convolutional features recurrently with the guidance of object boundary information. First, a deep convolutional network is used to extract multi-level feature maps and separately aggregate them into multiple resolutions, which can be used to generate coarse saliency maps. Meanwhile, another boundary information extraction branch is proposed to generate boundary features. Finally, an attention-based feature fusion module is designed to fuse boundary information into salient regions to achieve accurate boundary inference and semantic enhancement. The final saliency maps are the combination of the predicted boundary maps and integrated saliency maps, which are more closer to the ground truths. Experiments and analysis on four large-scale benchmarks verify that our framework achieves new state-of-the-art results.

3MNet: Multi-task, multi-level and multi-channel feature aggregation network for salient object detection

“…Through the collaborative feature learning of these two related tasks, the shared convolutional layer produces effective object perception features. Zhuge et al [20] focus on using the boundary features of the objects in the image, utilizing edge truth labels to supervise and refine the details of the detection feature map. [44] make full use of the multi-temporal features and show the effectiveness of multiple features in improving detection performance.…”

Learning to Detect Instance-level Salient Objects Using Complementary Image Labels

“…Xu et al [56] propose a CRF-based architecture to refine boundaries of both deep features and saliency maps in a coarse-to-fine manner. Some methods [74,71,53,42] propose to formulate saliency and edge detection with two network branches as multi-task learning. Feature fusion strategies have also been widely explored in salient object detection.…”

“…It often serves as a core step for downstream vision tasks like video object segmentation [50], object proposal generation [4], and image cropping [49]. Recent deep learningbased SOD methods have achieved a significant performance progress [47,74,56,42,68,17,48], benefited from the powerful representation learning capability of neural networks and large-scale pixel-level annotated training data. Since annotating pixel-level labels is extremely tedious, there are some works [47,60] that aim to explore cheaper image-level labels (e.g., class labels) to train SOD models in a weakly-supervised manner.…”

Multi-Modal Weights Sharing and Hierarchical Feature Fusion for RGBD Salient Object Detection

“…Liu and Han [29] introduced a hierarchical architecture which supervise the multi-scale coarse saliency maps with different scales ground truth maps to refine the details of saliency map hierarchically and progressively. Zhuge et al [30] provided stage-wise refinement frameworks to gradually enhance the boundary information. These methods make remarkable success.…”