As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
Recently, with the development of Convolutional Neural Networks (CNNs), deep learning-based saliency detection methods have advanced significantly. Most of the existing deep learning-based methods attempt to extract semantic context information to yield a saliency map. However, it is difficult to capture irregular context features by using a standard convolution because such features are often unevenly distributed in complex scenes. To address this problem, this paper proposes a novel saliency detection model named DCFA, which is implemented using two important modules. First, we design a Deformable Feature Extraction Module (DFEM) to focus on the unevenly distributed context features in both low-level details and high-level semantic information. Second, a Channel and Spatial Attention Module (CSAM) is devised to assign the adaptive weights of the features in the space and channel domains. The experimental results show that the proposed model can achieve the state-of-the-art performance on six widely used saliency detection benchmarks. Furthermore, our proposed network is end-to-end and runs at a speed of 20 fps on a single GPU.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.
This website uses cookies
We use cookies to provide you with the best possible experience. They also allow us to analyze user behavior in order to constantly improve the website for you. Info about the privacy policy of IOS Press.