arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1609.01064cs.CV
收藏

一种用于显著性预测的深度多层网络

A Deep Multi-Level Network for Saliency Prediction

Università degli Studi di Modena e Reggio Emilia(摩德纳和雷焦艾米利亚大学)

Marcella Cornia, Lorenzo Baraldi, Giuseppe Serra, Rita Cucchiara

摘要:

本文提出了一种新的显着性预测的深度架构。用于显著性预测的现有技术模型采用全卷积网络,该全卷积网络执行从最后一个卷积层提取的特征的非线性组合以预测显著性图。我们提出了一种架构,它结合了在卷积神经网络(CNN)的不同层次上提取的特征。我们的模型由三个主要模块组成:特征提取CNN,特征编码网络,对低级和高级特征图进行加权,以及先验学习网络。我们将我们的解决方案与两个公共基准数据集上的最新显着性模型进行了比较。结果表明,我们的模型在SALICON数据集的所有评估指标下都表现出色,SALICON数据集是目前最大的显着性预测公共数据集,并在MIT300基准测试中取得了有竞争力的结果。

英文摘要:

This paper presents a novel deep architecture for saliency prediction. Current state of the art models for saliency prediction employ Fully Convolutional networks that perform a non-linear combination of features extracted from the last convolutional layer to predict saliency maps. We propose an architecture which, instead, combines features extracted at different levels of a Convolutional Neural Network (CNN). Our model is composed of three main blocks: a feature extraction CNN, a feature encoding network, that weights low and high level feature maps, and a prior learning network. We compare our solution with state of the art saliency models on two public benchmarks datasets. Results show that our model outperforms under all evaluation metrics on the SALICON dataset, which is currently the largest public dataset for saliency prediction, and achieves competitive results on the MIT300 benchmark.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑