Search Results for author:
Found 69 papers, 48 papers with code
Leveraging Uncertainty for Deep Interpretable Classification and Weakly-Supervised Segmentation of Histology Images
Our method is composed of two networks: a localizer that yields segmentation mask, followed by a classifier.
Realistic Evaluation of Transductive Few-Shot Learning
Transductive inference is widely used in few-shot learning, as it leverages the statistics of the unlabeled query set of a few-shot task, typically yielding substantially better performances than its inductive counterpart.
Parameter-free Online Test-time Adaptation
An interesting and practical paradigm is online test-time adaptation, according to which training data is inaccessible, no labelled data from the test distribution is available, and adaptation can only happen at test time and on a handful of samples.
Negative Evidence Matters in Interpretable Histology Image Classification
The CNN is exploited to collect both positive and negative evidence at the pixel level to train the decoder.
The Devil is in the Margin: Margin-based Label Smoothing for Network Calibration
Following our observations, we propose a simple and flexible generalization based on inequality constraints, which imposes a controllable margin on logit distances.
Mixed-supervised segmentation: Confidence maximization helps knowledge distillation
In this work, we propose a dual-branch architecture, where the upper branch (teacher) receives strong annotations, while the bottom one (student) is driven by limited supervision and guided by the upper branch.
F-CAM: Full Resolution Class Activation Maps via Guided Parametric Upscaling
Interpolation is required to restore full size CAMs, yet it does not consider the statistical properties of objects, such as color and texture, leading to activations with inconsistent boundaries, and inaccurate localizations.
Source-Free Domain Adaptation for Image Segmentation
Our method yields comparable results to several state of the art adaptation techniques, despite having access to much less information, as the source images are entirely absent in our adaptation phase.
Mutual-Information Based Few-Shot Classification
We motivate our transductive loss by deriving a formal relation between the classification accuracy and mutual-information maximization.
Transductive Few-Shot Learning: Clustering is All You Need?
Surprisingly, we found that even standard clustering procedures (e. g., K-means), which correspond to particular, non-regularized cases of our general model, already achieve competitive performances in comparison to the state-of-the-art in few-shot learning.
Adversarial Robustness via Fisher-Rao Regularization
Adversarial robustness has become a topic of growing interest in machine learning since it was observed that neural networks tend to be brittle.
Beyond pixel-wise supervision for segmentation: A few global shape descriptors might be surprisingly good!
We also found that shape descriptors can be a valid way to encode anatomical priors about the task, enabling to leverage expert knowledge without additional annotations.
The hidden label-marginal biases of segmentation losses
In the abundant segmentation literature, there is no clear consensus as to which of these losses is a better choice, with varying performances for each across different benchmarks and applications.
Teach me to segment with mixed supervision: Confident students become masters
In conjunction with a standard cross-entropy over the labeled pixels, our novel formulation integrates two important terms: (i) a Shannon entropy loss defined over the less-supervised images, which encourages confident student predictions at the bottom branch; and (ii) a Kullback-Leibler (KL) divergence, which transfers the knowledge from the predictions generated by the strongly supervised branch to the less-supervised branch, and guides the entropy (student-confidence) term to avoid trivial solutions.
Few-Shot Segmentation Without Meta-Learning: A Good Transductive Inference Is All You Need?
We show that the way inference is performed in few-shot segmentation tasks has a substantial effect on performances -- an aspect often overlooked in the literature in favor of the meta-learning paradigm.
Annotation-efficient deep learning for automatic medical image segmentation
Automatic medical image segmentation plays a critical role in scientific research and medical care.
Information Maximization for Few-Shot Learning
We introduce Transductive Infomation Maximization (TIM) for few-shot learning.
Augmented Lagrangian Adversarial Attacks
Our attack enjoys the generality of penalty methods and the computational efficiency of distance-customized algorithms, and can be readily used for a wide set of distances.
Deep Interpretable Classification and Weakly-Supervised Segmentation of Histology Images via Max-Min Uncertainty
We propose novel regularization terms, which enable the model to seek both non-discriminative and discriminative regions, while discouraging unbalanced segmentations.
Deep Active Learning for Joint Classification & Segmentation with Weak Annotator
CNN visualization and interpretation methods, like class-activation maps (CAMs), are typically used to highlight the image regions linked to class predictions.
Cost-Sensitive Regularization for Diabetic Retinopathy Grading from Eye Fundus Images
Assessing the degree of disease severity in biomedical images is a task similar to standard classification but constrained by an underlying structure in the label space.
The Little W-Net That Could: State-of-the-Art Retinal Vessel Segmentation with Minimalistic Models
Our analysis demonstrates that the retinal vessel segmentation problem is far from solved when considering test images that differ substantially from the training data, and that this task represents an ideal scenario for the exploration of domain adaptation techniques.
Transductive Information Maximization For Few-Shot Learning
We introduce Transductive Infomation Maximization (TIM) for few-shot learning.
A Flow-Guided Mutual Attention Network for Video-Based Person Re-Identification
Our Mutual Attention network relies on the joint spatial attention between image and optical flow features maps to activate a common set of salient features across them.
Optical Flow Estimation
Video-Based Person Re-Identification
Medical Imaging with Deep Learning: MIDL 2020 -- Short Paper Track
This compendium gathers all the accepted extended abstracts from the Third International Conference on Medical Imaging with Deep Learning (MIDL 2020), held in Montreal, Canada, 6-9 July 2020.
Laplacian Regularized Few-Shot Learning
Our transductive inference does not re-train the base model, and can be viewed as a graph clustering of the query set, subject to supervision constraints from the support set.
Learning Data Augmentation with Online Bilevel Optimization for Image Classification
Data augmentation is a key practice in machine learning for improving generalization performance.
Source-Relaxed Domain Adaptation for Image Segmentation
Our formulation is based on minimizing a label-free entropy loss defined over target-domain data, which we further guide with a domain invariant prior on the segmentation regions.
Bounding boxes for weakly supervised segmentation: Global constraints get close to full supervision
Particularly, we leverage a classical tightness prior to a deep learning setting via imposing a set of constraints on the network outputs.
A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses
Second, we show that, more generally, minimizing the cross-entropy is actually equivalent to maximizing the mutual information, to which we connect several well-known pairwise losses.
Ranked #7 on
Metric Learning
on In-Shop
(using extra training data)
Semi-supervised few-shot learning for medical image segmentation
To handle this new learning paradigm, we propose to include surrogate tasks that can leverage very powerful supervisory signals --derived from the data itself-- for semantic feature learning.
On the Texture Bias for Few-Shot CNN Segmentation
Despite the initial belief that Convolutional Neural Networks (CNNs) are driven by shapes to perform visual recognition tasks, recent evidence suggests that texture bias in CNNs provides higher performing models when learning on large labeled training datasets.
Ranked #2 on
Few-Shot Semantic Segmentation
on Pascal5i
Non-parametric Uni-modality Constraints for Deep Ordinal Classification
We propose a new constrained-optimization formulation for deep ordinal classification, in which uni-modality of the label distribution is enforced implicitly via a set of inequality constraints over all the pairs of adjacent labels.
Ranked #1 on
Historical Color Image Dating
on HCI
Connectivity-constrained interactive annotations for panoptic segmentation
Large-scale ground truth data sets are of crucial importance for deep learning based segmentation models, but annotating per-pixel masks is prohibitively time consuming.
Adversarial Learning of General Transformations for Data Augmentation
Data augmentation (DA) is fundamental against overfitting in large convolutional neural networks, especially with a limited training dataset.
Deep Weakly-Supervised Learning Methods for Classification and Localization in Histology Images: A Comparative Study
Four key challenges are identified for the application of deep WSOL methods in histology -- under/over activation of CAMs, sensitivity to thresholding, and model selection.
Discretely-constrained deep network for weakly supervised segmentation
An efficient strategy for weakly-supervised segmentation is to impose constraints or regularization priors on target regions.
Universal Adversarial Audio Perturbations
We demonstrate the existence of universal adversarial perturbations, which can fool a family of audio classification architectures, for both targeted and untargeted attack scenarios.
Constrained domain adaptation for Image segmentation
We propose to adapt segmentation networks with a constrained formulation, which embeds domain-invariant prior knowledge about the segmentation regions.
Min-max Entropy for Weakly Supervised Pointwise Localization
Pointwise localization allows more precise localization and accurate interpretability, compared to bounding box, in applications where objects are highly unstructured such as in medical domain.
Exploiting Prunability for Person Re-Identification
Then, these techniques are analysed according to their pruningcriteria and strategy, and according to different scenarios for exploiting pruningmethods to fine-tuning networks to target domains.
Variational Fair Clustering
We derive a general tight upper bound based on a concave-convex decomposition of our fairness term, its Lipschitz-gradient property and the Pinsker's inequality.
Curriculum semi-supervised segmentation
This study investigates a curriculum-style strategy for semi-supervised CNN segmentation, which devises a regression network to learn image-level information such as the size of a target region.
Left Ventricle Segmentation
Semi-Supervised Semantic Segmentation
Constrained Deep Networks: Lagrangian Optimization via Log-Barrier Extensions
While sub-optimality is not guaranteed for non-convex problems, this result shows that log-barrier extensions are a principled way to approximate Lagrangian optimization for constrained CNNs via implicit dual variables.
On Direct Distribution Matching for Adapting Segmentation Networks
We juxtapose our approach to state-of-the-art segmentation adaptation via adversarial training in the network-output space.
Boundary loss for highly unbalanced segmentation
We propose a boundary loss, which takes the form of a distance metric on the space of contours, not regions.
Brain Lesion Segmentation From Mri
Ischemic Stroke Lesion Segmentation
+3
Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defenses
Research on adversarial examples in computer vision tasks has shown that small, often imperceptible changes to an image can induce misclassification, which has security implications for a wide range of image processing systems.
IVD-Net: Intervertebral disc localization and segmentation in MRI with a multi-modal UNet
Despite the technological advances in medical imaging, IVD localization and segmentation are still manually performed, which is time-consuming and prone to errors.
Scalable Laplacian K-modes
Furthermore, we show that the density modes can be obtained as byproducts of the assignment variables via simple maximum-value operations whose additional computational cost is linear in the number of data points.
Dense Multi-path U-Net for Ischemic Stroke Lesion Segmentation in Multiple Image Modalities
First, instead of combining the available image modalities at the input, each of them is processed in a different path to better exploit their unique information.
Deep clustering: On the link between discriminative models and K-means
Typically, they use multinomial logistic regression posteriors and parameter regularization, as is very common in supervised learning.
Ranked #2 on
Image Clustering
on YouTube Faces DB
Beyond Gradient Descent for Regularized Segmentation Losses
Both loss functions and architectures are often explicitly tuned to be amenable to this basic local optimization.
Multi-region segmentation of bladder cancer structures in MRI with progressive dilated convolutional networks
Precise segmentation of bladder walls and tumor regions is an essential step towards non-invasive identification of tumor stage and grade, which is critical for treatment decision and prognosis of patients with bladder cancer (BC).
Constrained-CNN losses for weakly supervised segmentation
To the best of our knowledge, the method of [Pathak et al., 2015] is the only prior work that addresses deep CNNs with linear constraints in weakly supervised segmentation.
Medical Image Segmentation
Weakly supervised segmentation
+1
HyperDense-Net: A hyper-densely connected CNN for multi-modal image segmentation
Therefore, the proposed network has total freedom to learn more complex combinations between the modalities, within and in-between all the levels of abstraction, which increases significantly the learning representation.
Ranked #1 on
Medical Image Segmentation
on iSEG 2017 Challenge
On Regularized Losses for Weakly-supervised CNN Segmentation
This approach simplifies weakly-supervised training by avoiding extra MRF/CRF inference steps or layers explicitly generating full masks, while improving both the quality and efficiency of training.
An ILP Solver for Multi-label MRFs with Connectivity Constraints
The method enforces connectivity priors iteratively by a cutting plane method, and provides feasible solutions with a guarantee on sub-optimality even if we terminate it earlier.
Deep CNN ensembles and suggestive annotations for infant brain MRI segmentation
We report evaluations of our method on the public data of the MICCAI iSEG-2017 Challenge on 6-month infant brain MRI segmentation, and show very competitive results among 21 teams, ranking first or second in most metrics.
Ranked #1 on
Infant Brain Mri Segmentation
on iSEG 2017 Challenge
Isointense Infant Brain Segmentation with a Hyper-dense Connected Convolutional Neural Network
Neonatal brain segmentation in magnetic resonance (MR) is a challenging problem due to poor image quality and low contrast between white and gray matter regions.
Kernel clustering: density biases and solutions
We call it Breiman's bias due to its similarity to the histogram mode isolation previously discovered by Breiman in decision tree learning with Gini impurity.
Unbiased Shape Compactness for Segmentation
We propose to constrain segmentation functionals with a dimensionless, unbiased and position-independent shape compactness prior, which we solve efficiently with an alternating direction method of multipliers (ADMM).
A 3D fully convolutional neural network and a random walker to segment the esophagus in CT
These figures translate into a very good agreement with the reference contours and an increase in accuracy compared to other methods.
DOPE: Distributed Optimization for Pairwise Energies
We formulate an Alternating Direction Method of Mul-tipliers (ADMM) that systematically distributes the computations of any technique for optimizing pairwise functions, including non-submodular potentials.
Secrets of GrabCut and Kernel K-Means
Our bound formulation for kernel K-means allows to combine general pair-wise feature clustering methods with image grid regularization using graph cuts, similarly to standard color model fitting techniques for segmentation.
Kernel Cuts: MRF meets Kernel & Spectral Clustering
We propose a new segmentation model combining common regularization energies, e. g. Markov Random Field (MRF) potentials, and standard pairwise clustering criteria like Normalized Cut (NC), average association (AA), etc.
Volumetric Bias in Segmentation and Reconstruction: Secrets and Solutions
Many standard optimization methods for segmentation and reconstruction compute ML model estimates for appearance or geometry of segments, e. g. Zhu-Yuille 1996, Torr 1998, Chan-Vese 2001, GrabCut 2004, Delong et al. 2012.
Submodularization for Binary Pairwise Energies
We propose a general optimization framework based on local submodular approximations (LSA).
Submodularization for Quadratic Pseudo-Boolean Optimization
We propose a general optimization framework based on local submodular approximations (LSA).
Auxiliary Cuts for General Classes of Higher Order Functionals
From these general-form bounds, we state various non-linear problems as the optimization of auxiliary functionals by graph cuts.
