527 citations · 1.1k across the 15 of their papers we have counts for
16 papers
Prompting with the human-touch: evaluating model-sensitivity of foundation models for musculoskeletal CT segmentation
Caroline Magg, Maaike A. ter Wee, Johannes G. G. Dobbe +4
Promptable Foundation Models (FMs), initially introduced for natural image segmentation, have also revolutionized medical image segmentation. The increasing number of models, along…
In-hoc Concept Representations to Regularise Deep Learning in Medical Imaging
Valentina Corbetta, Floris Six Dijkstra, Regina Beets-Tan +3
Deep learning models in medical imaging often achieve strong in-distribution performance but struggle to generalise under distribution shifts, frequently relying on spurious correl…
Zero-shot capability of SAM-family models for bone segmentation in CT scans
Caroline Magg, Hoel Kervadec, Clara I. Sánchez
The Segment Anything Model (SAM) and similar models build a family of promptable foundation models (FMs) for image and video segmentation. The object of interest is identified usin…
Leveraging point annotations in segmentation learning with boundary loss
Eva Breznik, Hoel Kervadec, Filip Malmberg +4
This paper investigates the combination of intensity-based distance maps with boundary loss for point-supervised semantic segmentation. By design the boundary loss imposes a strong…
On the dice loss gradient and the ways to mimic it
Hoel Kervadec, Marleen de Bruijne
In the past few years, in the context of fully-supervised semantic segmentation, several losses -- such as cross-entropy and dice -- have emerged as de facto standards to supervise…
Source-Free Domain Adaptation for Image Segmentation
Mathilde Bateson, Hoel Kervadec, Jose Dolz +2
Domain adaptation (DA) has drawn high interest for its capacity to adapt a model trained on labeled source data to perform well on unlabeled or weakly labeled target data from a di…
Beyond pixel-wise supervision for segmentation: A few global shape descriptors might be surprisingly good!
Hoel Kervadec, Houda Bahig, Laurent Letourneau-Guillon +2
Standard losses for training deep segmentation networks could be seen as individual classifications of pixels, instead of supervising the global shape of the predicted segmentation…
Few-Shot Segmentation Without Meta-Learning: A Good Transductive Inference Is All You Need?
Malik Boudiaf, Hoel Kervadec, Ziko Imtiaz Masud +3
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…
Source-Relaxed Domain Adaptation for Image Segmentation
Mathilde Bateson, Hoel Kervadec, Jose Dolz +2
Domain adaptation (DA) has drawn high interests for its capacity to adapt a model trained on labeled source data to perform well on unlabeled or weakly labeled target data from a d…
Bounding boxes for weakly supervised segmentation: Global constraints get close to full supervision
Hoel Kervadec, Jose Dolz, Shanshan Wang +2
We propose a novel weakly supervised learning segmentation based on several global constraints derived from box annotations. Particularly, we leverage a classical tightness prior t…
Discretely-constrained deep network for weakly supervised segmentation
Jizong Peng, Hoel Kervadec, Jose Dolz +3
An efficient strategy for weakly-supervised segmentation is to impose constraints or regularization priors on target regions. Recent efforts have focused on incorporating such cons…
Constrained domain adaptation for Image segmentation
Mathilde Bateson, Jose Dolz, Hoel Kervadec +2
We propose to adapt segmentation networks with a constrained formulation, which embeds domain-invariant prior knowledge about the segmentation regions. Such knowledge may take the…
Curriculum semi-supervised segmentation
Hoel Kervadec, Jose Dolz, Eric Granger +1
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…
Constrained Deep Networks: Lagrangian Optimization via Log-Barrier Extensions
Hoel Kervadec, Jose Dolz, Jing Yuan +3
This study investigates imposing hard inequality constraints on the outputs of convolutional neural networks (CNN) during training. Several recent works showed that the theoretical…
Boundary loss for highly unbalanced segmentation
Hoel Kervadec, Jihene Bouchtiba, Christian Desrosiers +3
Widely used loss functions for CNN segmentation, e.g., Dice or cross-entropy, are based on integrals over the segmentation regions. Unfortunately, for highly unbalanced segmentatio…
Constrained-CNN losses for weakly supervised segmentation
Hoel Kervadec, Jose Dolz, Meng Tang +3
Weakly-supervised learning based on, e.g., partially labelled images or image-tags, is currently attracting significant attention in CNN segmentation as it can mitigate the need fo…