Deep Level Sets for Salient Object Detection

By Ping Hu, Bing Shuai, Jun Liu, G. Wang. 2017

192 citations recorded by the source corpus.

Deep learning has been applied to saliency detection in recent years. The superior performance has proved that deep networks can model the semantic properties of salient objects. Yet it is difficult for a deep network to discriminate pixels belonging to similar receptive fields around the object boundaries, thus deep networks may output maps with blurred saliency and inaccurate boundaries. To tackle such an issue, in this work, we propose a deep Level Set network to produce compact and uniform saliency maps. Our method drives the network to learn a Level Set function for salient objects so it can output more accurate boundaries and compact saliency. Besides, to propagate saliency information among pixels and recover full resolution saliency map, we extend a superpixel-based guided filter to be a layer in the network. The proposed network has a simple structure and is trained end-to-end. During testing, the network can produce saliency maps by efficiently feedforwarding testing images at a speed over 12FPS on GPUs. Evaluations on benchmark datasets show that the proposed method achieves state-of-the-art performance.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

Citations in the corpus

Deep Level Sets: Implicit Surface Representations for 3D Shape Inference (2019, methodology): “Convolutional neural networks [9,12,16,22,28] and generative adversarial models (GANs) [3, 24] have been successfully applied to 3D reconstruction problems by using either volumetric or point cloud representations.”
Implicit Surface Representations As Layers in Neural Networks (2019, methodology): “Convolutional neural networks [9,12,16,22,29] and generative adversarial models (GANs) [3, 24] have been successfully applied to 3D reconstruction problems by using either volumetric or point cloud representations.”
Reverse Attention for Salient Object Detection (2018, methodology): “We compare the proposed method with 10 state-of-the-art ones, including 9 recent CNN-based approaches, DCL+ [22], DHS [26], SSD [16], RFCN [39], DLS [10], NLDF [30], DSS and DSS+ [8], Amulet [45], UCF [46], and one conventional top approach, DRFI [13], where symbol “+” indicates that the network inc”
PiCANet: Learning Pixel-Wise Contextual Attention for Saliency Detection (2017, methodology): “We also follow [5, 34, 10] to adopt it as one of our metrics with the default settings in [27].”
EGNet: Edge Guidance Network for Salient Object Detection (2019, methodology): “Besides, there are some methods using pre-processing (Superpixel) [20] or post-processing (CRF) [17, 28, 33] to preserve the object boundaries.”
Delving Deeper into Anti-Aliasing in ConvNets (2020, background): “ntent. Bilateral [25] and guided [10] ﬁlters are content aware as they can simultaneously preserve edge information while removing noise. Recent works integrate such classic ﬁlters into deep networks [14,35,37]. However, directly integrating these modules into a neural network requires careful tunin”
Object Instance Annotation With Deep Extreme Level Set Evolution (2019, methodology): “[21] adds the level set energy in the loss function and uses a CNN to directly predict the level set function for salient object detection.”
Salient Object Detection With Pyramid Attention and Salient Edges (2019, methodology): “We compare the proposed PAGE-Net against 19 recent deep learning based alternatives: MDF [21], LEGS [34], DS [24], DCL [22], ELD [20], MC [57], RFCN [36], DHS [26], HEDS [14], KSR [38], NLDF [29], DLS [15], AMU [54], UCF [55], SRM [37], FSN [8], PAGR [56], RAS [7] and C2S [23].”
Salient Object Detection in the Deep Learning Era: An In-Depth Survey (2019, methodology): “• DLS [75] utilizes a stack of convolution and dilated convolution layers to produce an initial saliency map, and then reﬁnes it at super-pixel level.”
DARNet: Deep Active Ray Network for Building Segmentation (2019, unknown): “[9, 23] propose to combine a CNN with a level set method [16] in an end-to-end differentiable manner.”
BASNet: Boundary-Aware Salient Object Detection (2019, background): “[18] proposed to learn a Level Set [48] function to output accurate boundaries and compact saliency.”
An Empirical Study of Language CNN for Image Captioning (2016, methodology): “Inspired by the recent success of CNNs in computer vision [10, 14], we adopt a language CNN with a hierarchical structure to capture the long-range dependencies be-tween the input words, called CNN L .”
Box-supervised Instance Segmentation with Level Set Evolution (2022, methodology): “Some works [15, 17, 21, 49, 54] have been proposed to embed the level set into the deep network in an end-to-end manner and achieve promising segmentation results.”
Spatiotemporal Knowledge Distillation for Efficient Estimation of Aerial Video Saliency (2019, methodology): “To address the feature issue, deep saliency models [35], [36], [37] proposed to use feature representations learned from data by using Convolutional Neural Networks (CNNs).”
Devil Is in the Edges: Learning Semantic Boundaries From Noisy Annotations (2019, methodology): “Level Set Methods [27] have been widely used for image segmentation [8, 13, 30, 18, 24, 5, 22, 14] due to their ability to automatically handle various topological changes such as splitting and merging.”