An ILP Model for Multi-Label MRFs With Connectivity Constraints

Shen, Ruobing; Bo, Tang; Lodi, Andrea; Tramontani, Andrea; Ayed, Ismail Ben

doi:10.1109/tip.2020.2995056

Cited by 4 publications

(2 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although our method can be used with any differentiable or non-differentiable constraint, in this paper, we illustrate it on a well-known constraint with broad applicability: connectivity. Given a segmented region G, we say that G is connected if and only if there exists a path between each pair of pixels p, q ∈ G such that all pixels in the path belong to G. Imposing connectivity in segmentation leads to a highly-complex problem which can only be solved for simplified cases, for example, by representing an image as a small set of superpixels (Shen et al, 2020). However, considering connectivity over the whole image may not be practical since it is hard to achieve in the early training stages.…”

Section: Local Connectivity Constraintsmentioning

confidence: 99%

Context-aware virtual adversarial training for anatomically-plausible segmentation

Wang

Peng

Pedersoli

et al. 2021

Preprint

View full text Add to dashboard Cite

Despite their outstanding accuracy, semi-supervised segmentation methods based on deep neural networks can still yield predictions that are considered anatomically impossible by clinicians, for instance, containing holes or disconnected regions.To solve this problem, we present a Context-aware Virtual Adversarial Training (CAVAT) method for generating anatomically plausible segmentation. Unlike approaches focusing solely on accuracy, our method also considers complex topological constraints like connectivity which cannot be easily modeled in a differentiable loss function. We use adversarial training to generate examples violating the constraints, so the network can learn to avoid making such incorrect predictions on new examples, and employ the REINFORCE algorithm to handle non-differentiable segmentation constraints. The proposed method offers a generic and efficient way to add any constraint on top of any segmentation network. Experiments on two clinically-relevant datasets show our method to produce segmentations that are both accurate and anatomically-plausible in terms of region connectivity.

show abstract

Section: Local Connectivity Constraintsmentioning

confidence: 99%

Context-aware virtual adversarial training for anatomically-plausible segmentation

Wang

Peng

Pedersoli

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The higher-order binary Markov random field (HoMRF) is a non-convex optimization model widely used in the fields of economy, information theory, quantum computing, machine learning and image analysis [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 ]. Recently, a new order reduction method has been developed to optimize HoMRF energies.…”

Section: Introductionmentioning

confidence: 99%

An Order Reduction Design Framework for Higher-Order Binary Markov Random Fields

Chen

Yang

2023

Entropy

View full text Add to dashboard Cite

The order reduction method is an important approach to optimize higher-order binary Markov random fields (HoMRFs), which are widely used in information theory, machine learning and image analysis. It transforms an HoMRF into an equivalent and easier reduced first-order binary Markov random field (RMRF) by elaborately setting the coefficients and auxiliary variables of RMRF. However, designing order reduction methods is difficult, and no previous study has investigated this design issue. In this paper, we propose an order reduction design framework to study this problem for the first time. Through study, we find that the design difficulty mainly lies in that the coefficients and variables of RMRF must be set simultaneously. Therefore, the proposed framework decomposes the design difficulty into two processes, and each process mainly considers the coefficients or auxiliary variables of RMRF. Some valuable properties are also proven. Based on our framework, a new family of 14 order reduction methods is provided. Experiments, such as synthetic data and image denoising, demonstrate the superiority of our method.

show abstract

Context-Aware Virtual Adversarial Training for Anatomically-Plausible Segmentation

Wang

Peng

Pedersoli

et al. 2021

Medical Image Computing and Computer Assisted Intervention – MICCAI 2021

View full text Add to dashboard Cite

An ILP Model for Multi-Label MRFs With Connectivity Constraints

Cited by 4 publications

References 33 publications

Context-aware virtual adversarial training for anatomically-plausible segmentation

Context-aware virtual adversarial training for anatomically-plausible segmentation

An Order Reduction Design Framework for Higher-Order Binary Markov Random Fields

Context-Aware Virtual Adversarial Training for Anatomically-Plausible Segmentation

Contact Info

Product

Resources

About