Learning Uncertain Convolutional Features for Accurate Saliency Detection

@article{Zhang2017LearningUC,
  title={Learning Uncertain Convolutional Features for Accurate Saliency Detection},
  author={Pingping Zhang and D. Wang and Huchuan Lu and Hongyu Wang and Baocai Yin},
  journal={2017 IEEE International Conference on Computer Vision (ICCV)},
  year={2017},
  pages={212-221}
}
Deep convolutional neural networks (CNNs) have delivered superior performance in many computer vision tasks. [] Key Method We achieve this via introducing a reformulated dropout (R-dropout) after specific convolutional layers to construct an uncertain ensemble of internal feature units. In addition, we propose an effective hybrid upsampling method to reduce the checkerboard artifacts of deconvolution operators in our decoder network.

Figures and Tables from this paper

Image saliency detection via multi-scale iterative CNN
TLDR
This work proposes a multi-scale iterative CNN for salient object detection, which has two complementary subnetworks at different spatial scales, which can estimate the saliency map much more efficiently.
Saliency Detection with Deformable Convolution and Feature Attention
TLDR
A novel saliency detection model named DCFA is proposed, which is implemented using a Deformable Feature Extraction Module (DFEM) to focus on the unevenly distributed context features in both low-level details and high-level semantic information.
SuperVAE: Superpixelwise Variational Autoencoder for Salient Object Detection
TLDR
This paper proposes a novel salient object detection framework using a superpixelwise variational autoencoder (SuperVAE) network and proposes a perceptual loss to take advantage from deep pre-trained CNNs to train the SuperVAE network.
Deep Embedding Features for Salient Object Detection
TLDR
This paper proposes a novel approach that transforms prior information into an embedding space to select attentive features and filter out outliers for salient object detection and proposes a Guided Filter Refinement Network (GFRN) to jointly optimize the predicted results and the learnable guidance maps.
Reverse Attention-Based Residual Network for Salient Object Detection
TLDR
A compact and efficient deep network with high accuracy for salient object detection by employing residual learning to refine it progressively by only learning the residual in each side-output, which can be achieved with few convolutional parameters, therefore leads to high compactness and high efficiency.
Residual Learning for Salient Object Detection
TLDR
This paper proposes a residual learning strategy and introduces to gradually refine the coarse prediction scale-by-scale to remedy the errors between coarse saliency map and scale-matching ground truth masks, and designs a fully convolutional network which does not need any post-processing.
Image Saliency Detection with Low-Level Features Enhancement
TLDR
A deep convolutional network with low-level feature enhanced for image saliency detection that outperforms the state-of-the-art approaches in both accuracy and efficiency.
...
...

References

SHOWING 1-10 OF 52 REFERENCES
Saliency Detection with Recurrent Fully Convolutional Networks
TLDR
This paper develops a new saliency model using recurrent fully convolutional networks (RFCNs) that is able to incorporate saliency prior knowledge for more accurate inference and enables the network to capture generic representations of objects for saliency detection.
Deep Contrast Learning for Salient Object Detection
  • Guanbin Li, Yizhou Yu
  • Computer Science, Environmental Science
    2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2016
TLDR
This paper proposes an end-to-end deep contrast network that significantly improves the state of the art in salient object detection and extracts segment-wise features very efficiently, and better models saliency discontinuities along object boundaries.
Saliency detection by multi-context deep learning
TLDR
This paper proposes a multi-context deep learning framework for salient object detection that employs deep Convolutional Neural Networks to model saliency of objects in images and investigates different pre-training strategies to provide a better initialization for training the deep neural networks.
Visual saliency based on multiscale deep features
  • Guanbin Li, Yizhou Yu
  • Computer Science
    2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2015
TLDR
This paper discovers that a high-quality visual saliency model can be learned from multiscale features extracted using deep convolutional neural networks (CNNs), which have had many successes in visual recognition tasks.
Online Tracking by Learning Discriminative Saliency Map with Convolutional Neural Network
TLDR
An online visual tracking algorithm by learning discriminative saliency map using Convolutional Neural Network using hidden layers of the network to improve target localization accuracy and achieve pixel-level target segmentation.
DeepSaliency: Multi-Task Deep Neural Network Model for Salient Object Detection
TLDR
This paper proposes a multi-task deep saliency model based on a fully convolutional neural network with global input (whole raw images) and global output (Whole saliency maps) and presents a graph Laplacian regularized nonlinear regression model for saliency refinement.
DeepSaliency : MultiTask Deep Neural Network Model for Salient Object Detection
TLDR
This paper proposes a multi-task deep saliency model based on a fully convolutional neural network (FCNN) with global input (whole raw images) and global output (Whole saliency maps) and presents a graph Laplacian regularized nonlinear regression model for saliency refinement.
Deep Saliency with Encoded Low Level Distance Map and High Level Features
TLDR
It is demonstrated that hand-crafted features can provide complementary information to enhance performance of saliency detection that utilizes only high level features.
Learning Deconvolution Network for Semantic Segmentation
TLDR
A novel semantic segmentation algorithm by learning a deep deconvolution network on top of the convolutional layers adopted from VGG 16-layer net, which demonstrates outstanding performance in PASCAL VOC 2012 dataset.
Fully Convolutional Networks for Semantic Segmentation
TLDR
It is shown that convolutional networks by themselves, trained end- to-end, pixels-to-pixels, improve on the previous best result in semantic segmentation.
...
...