Effective Two-Stage Image Segmentation: A New Non-Lipschitz Decomposition Approach with Convergent Algorithm

Guo, Xifeng; Xue, Yunhua; Wu, Chunlin

doi:10.1007/s10851-020-01001-3

Cited by 5 publications

(3 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Following [8], people considered various decomposition models. Usually there is one cartoon component regularized by the TV (total variation) prior [8,46] or non-convex TV prior [9,13,23]. The TV prior is a composition of methods were proposed to complete segmentation task.…”

Section: 3mentioning

confidence: 99%

“…2b, our IDmUNet consists of two cascading stages: the first one employs a group of IDNets to extract meaningful feature layers and the second one uses a mini-UNet to operate segmentation on them. We mention that this structure is inspired by recent variational two-stage segmentation methods [4,5,9,23]. Our variational model in this papers is different from theirs.…”

Section: Idmunetmentioning

confidence: 99%

“…-norm and gradient operators, characterizing the sparsity[7,11,12] of an image in gradient domain. The other components depend on applications, such as cartoontexture decomposition[24,29,44], Retinex illumination estimation[33,38,41,42] and intensity inhomogeneity removal[4,5,9,23]. Also, in[4,5,9,23] two-stage…”

mentioning

confidence: 99%

See 2 more Smart Citations

IDmUNet: A new image decomposition induced network for sparse feature segmentation

Ren¹,

Gao²,

Wu³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

UNet and its variants are among the most popular methods for medical image segmentation. Despite their successes in task generality, most of them consider little mathematical modeling behind specific applications. In this paper, we focus on the sparse feature segmentation task and make a taskoriented network design, in which the target objects are sparsely distributed and the background is hard to be mathematically modeled. We start from an image decomposition model with sparsity regularization, and propose a deep unfolding network, namely IDNet, based on an iterative solver, scaled alternating direction method of multipliers (scaled-ADMM). The IDNet splits raw inputs into double feature layers. Then a new task-oriented segmentation network is constructed, dubbed as IDmUNet, based on the proposed IDNets and a mini-UNet. Because of the sparsity prior and deep unfolding method in the structure design, this IDmUNet combines the advantages of mathematical modeling and data-driven approaches. Firstly, our approach has mathematical interpretability and can achieve favorable performance with far fewer learnable parameters. Secondly, our IDmUNet is robust in a simple end-to-end training with explainable behaviors. In the experiments of retinal vessel segmentation (RVS), IDmUNet produces the state-of-the-art results with only 0.07m parameters, whereas SA-UNet, one of the latest variants of UNet, contains 0.54m and the original UNet 31.04m. Moreover, the training procedure of our network converges faster without overfitting phenomenon. This decomposition-based network construction strategy can be generalized to other problems with mathematically clear targets and complicated unclear backgrounds.

show abstract