Salient object detection via multi-scale attention CNN

Ji, Yuzhu; Zhang, Haijun; Wu, Q. M. Jonathan

doi:10.1016/j.neucom.2018.09.061

Cited by 82 publications

(27 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The attention mechanism is prominent for its ability to select discriminative features, and has been applied to many computer vision tasks, such as saliency detection [53]- [55], image captioning [56], image classification [57], semantic segmentation [58], image deblurring [59], and visual pose estimation [60].…”

Section: Attention Modelsmentioning

confidence: 99%

Co-Saliency Detection With Co-Attention Fully Convolutional Network

Gao

Zhao²,

Liu

et al. 2021

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Co-saliency detection aims to detect common salient objects from a group of relevant images. Some attempts have been made with the Fully Convolutional Network (FCN) framework and achieve satisfactory detection results. However, due to stacking convolution layers and pooling operation, the boundary details tend to be lost. In addition, existing models often utilize the extracted features without discrimination, leading to redundancy in representation since actually not all features are helpful to the final prediction and some even bring distraction. In this paper, we propose a co-attention module embedded FCN framework, called as Co-Attention FCN (CA-FCN). Specifically, the co-attention module is plugged into the high-level convolution layers of FCN, which can assign larger attention weights on the common salient objects and smaller ones on the background and uncommon distractors to boost final detection performance. Extensive experiments on three popular co-saliency benchmark datasets demonstrate the superiority of the proposed CA-FCN, which outperforms state-of-the-arts in most cases. Besides, the effectiveness of our new co-attention module is also validated with ablation studies.

show abstract

Section: Attention Modelsmentioning

confidence: 99%

Co-Saliency Detection With Co-Attention Fully Convolutional Network

Gao

Zhao²,

Liu

et al. 2021

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

show abstract

“…Recent years have witnessed the boom of deep convolutional nerual network in many challenging tasks, ranging from image classification to object detection [24,25,26,27,28]. However, it is computationally expensive because the amount of the computation scales increases linearly with the number of image pixels.…”

Section: Attention Mechanismmentioning

confidence: 99%

Attend to count: Crowd counting with adaptive capacity multi-scale CNNs

Zou

Cheng

et al. 2019

Neurocomputing

View full text Add to dashboard Cite

Crowd counting is a challenging task due to the large variations in crowd distributions. Previous methods tend to tackle the whole image with a single fixed structure, which is unable to handle diverse complicated scenes with different crowd densities. Hence, we propose the Adaptive Capacity Multi-scale convolutional neural networks (ACM-CNN), a novel crowd counting approach which can assign different capacities to different portions of the input. The intuition is that the model should focus on important regions of the input image and optimize its capacity allocation conditioning on the crowd intensive degree. ACM-CNN consists of three types of modules: a coarse network, a fine network, and a smooth network. The coarse network is used to explore the areas that need to be focused via count attention mechanism, and generate a rough feature map. Then the fine network processes the areas of interest into a fine feature map. To alleviate the sense of division caused by fusion, the smooth network is designed to combine two feature maps organically to produce high-quality density maps. Extensive experiments are conducted on five mainstream datasets. The results demonstrate the effectiveness of the proposed model for both density estimation and crowd counting tasks.

show abstract

“…Specifically, we attempt to exploit the attentive weights of each entity generated via the attention mechanism to achieve entity localization. The ability of the attention mechanism to select focusing regions by generating appropriate attentive weights has been proven in many previous studies . Hence, the attention process in the AWRNN can be regarded as estimating the locations of entities.…”

Section: Scene Graph Predictionmentioning

confidence: 99%

“…This idea has been proven to be advantageous for filtering out noisy patches and selecting the relevant ones in visual feature maps in an unsupervised manner. Considerable progress has been achieved by this attention property in solving visual tasks, including image captioning and object detection . Inspired by these results, we exploit the visual attention mechanism to automatically locate the target regions in visual space and dynamically output attentive features for the targets.…”

Section: Introductionmentioning

confidence: 99%

“…Considerable progress has been achieved by this attention property in solving visual tasks, including image captioning [25][26][27][28][29] and object detection. [30][31][32] Inspired by these results, we exploit the visual attention mechanism to automatically locate the target regions in visual space and dynamically output attentive features for the targets. To satisfy the second requirement, we treat the relationship triplet recognition as a sequential learning problem with a recurrent neural network (RNN).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A hierarchical recurrent approach to predict scene graphs from a visual‐attention‐oriented perspective

Gao

Zhu

Zhang

et al. 2019

Computational Intelligence

View full text Add to dashboard Cite

A scene graph provides a powerful intermediate knowledge structure for various visual tasks, including semantic image retrieval, image captioning, and visual question answering. In this paper, the task of predicting a scene graph for an image is formulated as two connected problems, ie, recognizing the relationship triplets, structured as , and constructing the scene graph from the recognized relationship triplets. For relationship triplet recognition, we develop a novel hierarchical recurrent neural network with visual attention mechanism. This model is composed of two attention‐based recurrent neural networks in a hierarchical organization. The first network generates a topic vector for each relationship triplet, whereas the second network predicts each word in that relationship triplet given the topic vector. This approach successfully captures the compositional structure and contextual dependency of an image and the relationship triplets describing its scene. For scene graph construction, an entity localization approach to determine the graph structure is presented with the assistance of available attention information. Then, the procedures for automatically converting the generated relationship triplets into a scene graph are clarified through an algorithm. Extensive experimental results on two widely used data sets verify the feasibility of the proposed approach.

show abstract

Salient object detection via multi-scale attention CNN

Cited by 82 publications

References 5 publications

Co-Saliency Detection With Co-Attention Fully Convolutional Network

Co-Saliency Detection With Co-Attention Fully Convolutional Network

Attend to count: Crowd counting with adaptive capacity multi-scale CNNs

A hierarchical recurrent approach to predict scene graphs from a visual‐attention‐oriented perspective

Contact Info

Product

Resources

About