DNA: Deeply Supervised Nonlinear Aggregation for Salient Object Detection

doi:10.1109/TCYB.2021.3051350

DNA: Deeply Supervised Nonlinear Aggregation for Salient Object Detection

2022-07-01

61

OA

AI

DOI：10.1109/TCYB.2021.3051350

Original

Save

View PDF

Abstract

En 中文

Recent progress on salient object detection mainly aims at exploiting how to effectively integrate multiscale 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 article, 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: 1) aggregates side-output features rather than predictions and 2) adopts nonlinear instead of linear transformations. Experiments demonstrate that DNA can successfully break through the bottleneck of the 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.

Keywords:

DNA

Saliency detection

Object detection

Feature extraction

Convolution

Detectors

Semantics

Deeply supervised nonlinear aggregation (DNA)

saliency detection

salient object detection

DNA: Deeply Supervised Nonlinear Aggregation for Salient Object Detection

Abstract

AI Summary

Journal

Organization