MSB-FCN: Multi-Scale Bidirectional FCN for Object Skeleton Extraction

Yang, Fan; Xin, Li; Shen, Jianbing

doi:10.1109/tip.2020.3038483

Cited by 21 publications

(27 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the first task, we considerŷ fuse as the results and compare them with five other leading methods on this field: HED, 12 SRN, 10 RCF, 20 HiFi, 21 and MSB-FCN. 13 For the regression task,Ŷ R is compared with Maxim's 9 and VGG-16, 15 which is used as the baseline in our experiments. Due to the final goal of our task is to get the locations of midline, we introduce the following equation to convert the probability map from skeleton extractionbased methods into coordinatesŶ midline to fairly compare with regression-based methods:…”

Section: C Experimental Resultsmentioning

confidence: 99%

“…Then the MSBI module inspired by Fan et al 13 is proposed to integrate all features from different scales to overcome semantic vs resolution conflict: low‐level features focus on detailed structures with high resolution while high‐level features focus on conceptual semantics with low resolution 11 . This module receives the outputs produced by the down‐sampling branch and generate

\overset{false^}{Y} \in R^{H \times W \times S}

.…”

Section: Methodsmentioning

confidence: 99%

“…Then F Á ð Þ is adopted to integrate features and keep the channels consistent with the ones in the current scale, which is different from the direct addition operation. 13 Finally, the additional convolution and ReLU are used to update h SD f i to get a higher representation.…”

Section: B Multiscale Line Detectionmentioning

confidence: 99%

“…12 Besides, the bottom-to-up manner can be utilized to equip the shallow layers (high resolution) with the ability to perceive high-level concepts. 10 In this paper, we introduce a novel multiscale bidirectional integration module (MSBI) inspired by Fan's method 13 to integrate semantics and details more effectively than others. Different from the fusing solutions mentioned above in Fig.…”

Section: Introductionmentioning

confidence: 99%

“…In this paper, we introduce a novel multiscale bidirectional integration module (MSBI) inspired by Fan's method 13 to integrate semantics and details more effectively than others. Different from the fusing solutions mentioned above in Fig.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

The delineation of largely deformed brain midline using regression‐based line detection network

et al. 2020

View full text Add to dashboard Cite

Purpose The human brain has two cerebral hemispheres that are roughly symmetric and separated by a midline, which is nearly a straight line shown in axial computed tomography (CT) images in healthy subjects. However, brain diseases such as hematoma and tumors often cause midline shift, where the degree of shift can be regarded as a quantitative indication in clinical practice. To facilitate clinical evaluation, we need computer‐aided methods to automate this quantification. Nevertheless, most existing studies focused on the landmark‐ or symmetry‐based methods that provide only the existence of shift or its maximum distance, which could be easily affected by anatomical variability and large brain deformations. Intuitive results such as midline delineation or measurement are lacking. In this study, we focus on developing an automated and robust method based on the fully convolutional neural network for the delineation of midline in largely deformed brains. Methods We propose a novel regression‐based line detection network (RLDN) for the robust midline delineation, especially in largely deformed brains. Specifically, to improve the robustness of delineation in largely deformed brains, we regard the delineation of the midline as the skeleton extraction task and then use the multiscale bidirectional integration module to acquire more representative features. Based on the skeleton extraction, we incorporate the regression task into it to delineate more accurate and continuous midline, especially in largely deformed brains. Our study utilized the public CQ 500 dataset (128 subjects) for training with hold‐out validation on 61 subjects from a private cohort accrued from a local hospital. Results The mean line distance error and F1‐score were 1.17 ± 0.72 mm with 0.78 on CQ 500 test set, and 4.15 ± 3.97 mm with 0.61 on the private dataset. Besides, significant differences (P < 0.05) were observed between our method and other comparative ones on these two datasets. Conclusions This work provides a novel solution to acquire robust delineation of the midline, especially in largely deformed brains, and achieves state‐of‐the‐art performance on the public and our private dataset, which makes it possible for automated diagnosis of relevant brain diseases in the future.

show abstract

Section: C Experimental Resultsmentioning

confidence: 99%

\overset{false^}{Y} \in R^{H \times W \times S}

.…”

Section: Methodsmentioning

confidence: 99%

Section: B Multiscale Line Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The delineation of largely deformed brain midline using regression‐based line detection network

et al. 2020

View full text Add to dashboard Cite

show abstract

Cascade Graph Neural Networks for RGB-D Salient Object Detection

Luo

Li²,

Yang³

et al. 2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

In this paper, we study the problem of salient object detection (SOD) for RGB-D images using both color and depth information. A major technical challenge in performing salient object detection from RGB-D images is how to fully leverage the two complementary data sources. Current works either simply distill prior knowledge from the corresponding depth map for handling the RGB-image or blindly fuse color and geometric information to generate the coarse depth-aware representations, hindering the performance of RGB-D saliency detectors. In this work, we introduce Cascade Graph Neural Networks (Cas-Gnn), a unified framework which is capable of comprehensively distilling and reasoning the mutual benefits between these two data sources through a set of cascade graphs, to learn powerful representations for RGB-D salient object detection. Cas-Gnn processes the two data sources individually and employs a novel Cascade Graph Reasoning (CGR) module to learn powerful dense feature embeddings, from which the saliency map can be easily inferred. Contrast to the previous approaches, the explicitly modeling and reasoning of high-level relations between complementary data sources allows us to better overcome challenges such as occlusions and ambiguities. Extensive experiments demonstrate that Cas-Gnn achieves significantly better performance than all existing RGB-D SOD approaches on several widely-used benchmarks. Code is available at https://github.com/LA30/Cas-Gnn.

show abstract