Incorporating prior knowledge into a segmentation task, whether it be under the form of geometrical constraints (area/volume penalisation, convexity enforcement, etc.) or of topological constraints (to preserve the co...
详细信息
Incorporating prior knowledge into a segmentation task, whether it be under the form of geometrical constraints (area/volume penalisation, convexity enforcement, etc.) or of topological constraints (to preserve the contextual relations between objects, to monitor the number of connected components), proves to increase accuracy in medical image segmentation. In particular, it allows to compensate for the issue of weak boundary definition, of imbalanced classes, and to be more in line with anatomical consistency even though the data do not explicitly exhibit those features. This observation underpins the introduced contribution that aims, in a hybrid setting, to leverage the best of both worlds that variationalmethods and supervised deep learning approaches embody: (a) versatility and adaptability in the mathematical formulation of the problem to encode geometrical/topological constraints, (b) interpretability of the results for the former formalism, while (c) more efficient and effective processing models, (d) ability to become more proficient at learning intricate features and executing more computationally intensive tasks, for the latter one. To be more precise, a unified variational framework involving topological prescriptions in the training of convolutional neural networks through the design of a suitable penalty in the loss function is provided. These topological constraints are implicitly enforced by viewing the segmentation procedure as a registration task between the processed image and its associated ground truth under incompressibility conditions, thus making them homeomorphic. A very preliminary version (Lambert et al., in Calatroni, Donatelli, Morigi, Prato, Santacesaria (eds) scalespace and variationalmethods in computervision, Springer, Berlin, 2023, pp. 363-375) of this work has been published in the proceedings of the Ninth internationalconference on scalespace and variationalmethods in computervision, 2023. It contained neither all the theo
Understanding the decision-making and trusting the reliability of Deep Machine Learning Models is crucial for adopting such methods to safety-relevant applications which play an important role in the digitization of t...
详细信息
ISBN:
(数字)9783031687389
ISBN:
(纸本)9783031687372;9783031687389
Understanding the decision-making and trusting the reliability of Deep Machine Learning Models is crucial for adopting such methods to safety-relevant applications which play an important role in the digitization of the railway system. We extend self-explainable Prototypical variational models with autoencoder-based Out-of-Distribution (OOD) detection: A variational Autoencoder is applied to learn a meaningful latent space which can be used for distance-based classification, likelihood estimation for OOD detection, and reconstruction. The In-Distribution (ID) region is defined by a Gaussian mixture distribution with learned prototypes representing the center of each mode. Furthermore, a novel restriction loss is introduced that promotes a compact ID region in the latent space without collapsing it into single points. The reconstructive capabilities of the Autoencoder ensure the explainability of the prototypes and the ID region of the classifier, further aiding the discrimination of OOD samples. Extensive evaluations on common OOD detection benchmarks as well as a large-scale dataset from a real-world railway application demonstrate the usefulness of the approach, outperforming previous methods.
We propose a model-driven neural fields approach for solving variational problems. The approach can be applied to a variety of problems with convex, 1-homogeneous regularizer and arbitrary, possibly non-convex, data t...
详细信息
Probabilistic diffusion models enjoy increasing popularity in the deep learning community. They generate convincing samples from a learned distribution of input images with a wide field of practical applications. Orig...
详细信息
Medical Visual Question Answering (Med-VQA) is pivotal for interpreting medical queries via corresponding images. While multi-modal fusion stages in Med-VQA benefit from attention mechanisms and Transformer-based meth...
详细信息
ISBN:
(纸本)9783031723520;9783031723537
Medical Visual Question Answering (Med-VQA) is pivotal for interpreting medical queries via corresponding images. While multi-modal fusion stages in Med-VQA benefit from attention mechanisms and Transformer-based methods, the latter's computational demands limit scalability. Emerging as a robust alternative, State space Models (SSMs), particularly the Mamba model, have shown promise in sequence modeling and deep learning network building. However, their limitation to single-modality data processing curtails their direct application in complex vision-language tasks inherent in Med-VQA. Additionally, we identify and address the underutilization of multi-scale visual information in existing Med-VQA frameworks, incorporating it into our fusion process for enhanced context comprehension. Our approach also features an innovative asymmetric fusion structure tailored to bridge the gap between open-ended and close-ended questions, optimizing question answering accuracy. Comparative analyses on benchmark datasets VQA-RAD and SLAKE underscore our method's efficiency, outperforming state-of-the-art Med-VQA models in accuracy while operating with significantly fewer parameters than Transformer-based counterparts. This study not only proposed a powerful Med-VQA model but also broadens the scope of SSMs in tackling complex multi-modal challenges.
The total generalized variation extends the total variation by incorporating higher-order smoothness. Thus, it can also suffer from similar discretization issues related to isotropy. Inspired by the success of novel d...
详细信息
Faces must be present in images for intelligent vision-based human computer interaction to work. Since many years ago, face recognition research has been ongoing and significant. This procedure entails facial tracking...
详细信息
We propose a learning paradigm for the numerical approximation of differential invariants of planar curves. Deep neural-networks’ (DNNs) universal approximation properties are utilized to estimate geometric measures....
详细信息
The proceedings contain 63 papers. The special focus in this conference is on scalespace and variationalmethods in computervision. The topics include: Fast Inexact Bilevel Optimization for Analytical Deep Imag...
ISBN:
(纸本)9783031923654
The proceedings contain 63 papers. The special focus in this conference is on scalespace and variationalmethods in computervision. The topics include: Fast Inexact Bilevel Optimization for Analytical Deep Image Priors;why do we Regularise in Every Iteration for Imaging Inverse Problems?;on an Analytical Inversion Formula for the Modulo Radon Transform;multiResolution Low-Rank Regularization of Dynamic Imaging Problems;a Fractional Graph La+Ψ Approach to Image Reconstruction;a Fractional-Order Telegraph Diffusion Model for Multiplicative Noise Removal;self-supervised Conformal Prediction for Uncertainty Quantification in Imaging Problems;real-Time Scene Recovery from Image scalespace and Perceptual Hue Similarity;a Novel Interpretation of the Radon Transform’s Ray and Pixel-Driven Discretizations Under Balanced Resolutions;Product of Gaussian Mixture Diffusion Model for Non-linear MRI Inversion;Whiteness-Based Bilevel Estimation of Weighted TV Parameter Maps for Image Denoising;Direct Atomistic Reconstruction in Homogeneous Cryo-EM Using Protein Geometry Regularization;enhanced Denoising and Convergent Regularisation Using Tweedie Scaling;max-Sparsity Atomic Autoencoders with Application to Inverse Problems;equivariant Bootstrap for Uncertainty Quantification in Image Classification;equivariant Denoisers for Image Restoration;max-Normalized Radon Cumulative Distribution Transform for Limited Data Classification;prediction of Parametric Surfaces for Multi-object Segmentation in 3D Biological Imaging;plug-and-Play Half-Quadratic Splitting for Ptychography;deep Unrolling for Learning Optimal Spatially Varying Regularisation Parameters for Total Generalised Variation;identifying Memorization of Diffusion Models Through p-Laplace Analysis;learning Anisotropic Metrics for Geodesic Distances via the Heat Equation for Image Segmentation;deceptive Diffusion: Generating Synthetic Adversarial Examples;TomoSelfDEQ: Self-supervised Deep Equilibrium Learning for Sparse-Angle C
The proceedings contain 63 papers. The special focus in this conference is on scalespace and variationalmethods in computervision. The topics include: Fast Inexact Bilevel Optimization for Analytical Deep Imag...
ISBN:
(纸本)9783031923685
The proceedings contain 63 papers. The special focus in this conference is on scalespace and variationalmethods in computervision. The topics include: Fast Inexact Bilevel Optimization for Analytical Deep Image Priors;why do we Regularise in Every Iteration for Imaging Inverse Problems?;on an Analytical Inversion Formula for the Modulo Radon Transform;multiResolution Low-Rank Regularization of Dynamic Imaging Problems;a Fractional Graph La+Ψ Approach to Image Reconstruction;a Fractional-Order Telegraph Diffusion Model for Multiplicative Noise Removal;self-supervised Conformal Prediction for Uncertainty Quantification in Imaging Problems;real-Time Scene Recovery from Image scalespace and Perceptual Hue Similarity;a Novel Interpretation of the Radon Transform’s Ray and Pixel-Driven Discretizations Under Balanced Resolutions;Product of Gaussian Mixture Diffusion Model for Non-linear MRI Inversion;Whiteness-Based Bilevel Estimation of Weighted TV Parameter Maps for Image Denoising;Direct Atomistic Reconstruction in Homogeneous Cryo-EM Using Protein Geometry Regularization;enhanced Denoising and Convergent Regularisation Using Tweedie Scaling;max-Sparsity Atomic Autoencoders with Application to Inverse Problems;equivariant Bootstrap for Uncertainty Quantification in Image Classification;equivariant Denoisers for Image Restoration;max-Normalized Radon Cumulative Distribution Transform for Limited Data Classification;prediction of Parametric Surfaces for Multi-object Segmentation in 3D Biological Imaging;plug-and-Play Half-Quadratic Splitting for Ptychography;deep Unrolling for Learning Optimal Spatially Varying Regularisation Parameters for Total Generalised Variation;identifying Memorization of Diffusion Models Through p-Laplace Analysis;learning Anisotropic Metrics for Geodesic Distances via the Heat Equation for Image Segmentation;deceptive Diffusion: Generating Synthetic Adversarial Examples;TomoSelfDEQ: Self-supervised Deep Equilibrium Learning for Sparse-Angle C
暂无评论