Laurent Najman

Image classification in specialized domains is often constrained by limited annotations, hindering the learning of effective feature representations. By coupling convolutional encoders with multilayer perceptrons (MLPs), conventional convolutional neural networks (CNNs) jointly learn feature representations and parametric decision boundaries, a process that becomes challenging in data-scarce settings, particularly when class distributions exhibit irregular geometric structures. Graph-based classifiers provide a complementary non-parametric alternative by exploiting structural relationships among samples, yet their effectiveness depends on meaningful Euclidean distances for arc-weight estimation, which standard CNNs do not explicitly encourage. To address this gap, we introduce Optimum-Path Forest Network (OPFNet), an architecture that replaces the typical MLP head of CNNs with Optimum-Path Forest (OPF), a non-parametric graph-based framework that exploits optimum-path connectivity to assign labels from a set of representative prototypes. OPFNet employs a tailored loss function to align the learned feature space with the Euclidean distance structure required for arc-weight computation. This alignment improves the suitability of the feature space geometry for connectivity-based decision making, particularly under limited supervision. By exploring different encoder initialization strategies and arc-weight estimation schemes, we demonstrate the flexibility of OPFNet across diverse configurations and supervision settings. Under limited supervision, the proposed architecture generally outperforms CNN and embedding-based baselines, while remaining competitive under moderate supervision across biomedical and remote sensing datasets.

In the framework of edge-weighted graphs, watersheds have proven to be linked to well-known optimization problems, as Minimum Spanning Tree, which allowed the design of efficient algorithms for computing (hierarchical) watershed segmentations. In the present article, after reviewing the literature related to watershed segmentation, we present a detailed end-to-end pipeline of algorithms to compute (hierarchical) watershed segmentations, starting from the computation of graph-based image representations, up to the computation of connected components of the final (hierarchical) segmentation. We consider the several variations of watersheds, including their supervised and unsupervised versions, and the various ways of computing seeds, to name a few. For the first time, we bring together all these watershed notions and algorithms in a compact and understandable way. We aim at providing a reference for those interested in employing and reimplementing the watershed segmentation framework for their task at hand.

After Learning Evolutionary Retrofitting (AfterLearnER) consists in applying evolutionary optimization to refine fully trained machine learning models by optimizing a set of carefully chosen parameters or hyperparameters of the model, with respect to some actual, exact, and hence possibly non-differentiable error signal, performed on a subset of the standard validation set. The efficiency of AfterLearnER is demonstrated by tackling non-differentiable signals such as threshold-based criteria in depth sensing, the word error rate in speech resynthesis, the number of kills per life at Doom, computational accuracy or BLEU in code translation, image quality in 3D generative adversarial networks (GANs), and user feedback in image generation via latent diffusion models (LDM). This retrofitting can be done after training, or dynamically at inference time by taking into account the user feedback. The advantages of AfterLearnER are its versatility, the possibility to use non-differentiable feedback, including human evaluations (i.e., no gradient is needed), the limited overfitting supported by a theoretical study, and its anytime behavior. Last but not least, AfterLearnER requires only a small amount of feedback, i.e., a few dozen to a few hundred scalars, compared to the tens of thousands needed in most related published works.

Salient Object Detection (SOD) methods can locate objects that stand out in an image, assign higher values to their pixels in a saliency map, and binarize the map outputting a predicted segmentation mask. A recent tendency is to investigate pre-trained lightweight models rather than deep neural networks in SOD tasks, coping with applications under limited computational resources. In this context, we have investigated lightweight networks using a methodology named Feature Learning from Image Markers (FLIM), which assumes that the encoder's kernels can be estimated from marker pixels on discriminative regions of a few representative images. This work proposes flyweight networks, hundreds of times lighter than lightweight models, for SOD by combining a FLIM encoder with an adaptive decoder, whose weights are estimated for each input image by a given heuristic function. Such FLIM networks are trained from three to four representative images only and without backpropagation, making the models suitable for applications under labeled data constraints as well. We study five adaptive decoders; two of them are introduced here. Differently from the previous ones that rely on one neuron per pixel with shared weights, the heuristic functions of the new adaptive decoders estimate the weights of each neuron per pixel. We compare FLIM models with adaptive decoders for two challenging SOD tasks with three lightweight networks from the state-of-the-art, two FLIM networks with decoders trained by backpropagation, and one FLIM network whose labeled markers define the decoder's weights. For one of the applications, we evaluate the generalization ability of the networks to six different datasets. The experiments demonstrate the advantages of the proposed networks over the baselines, revealing the importance of further investigating such methods in new applications.

Crack detection is an important task in computer vision. Despite impressive in-dataset performance, deep learning-based methods still struggle in generalizing to unseen domains. The thin structure property of cracks is usually overlooked by previous methods. In this work, we introduce CrackCue, a novel method for robust crack detection based on coarse-to-fine crack cue generation. The core concept lies on leveraging the thin structure property to generate a robust crack cue, guiding the crack detection. Specifically, we first employ a simple max-pooling and upsampling operation on the crack image. This results in a coarse crack-free background, based on which a fine crack-free background can be obtained via a reconstruction network. The difference between the original image and fine crack-free background provides a fine crack cue. This fine cue embeds robust crack prior information which is unaffected by complex backgrounds, shadow, and varied lighting. As a plug-and-play method, we incorporate the proposed CrackCue into three advanced crack detection networks. Extensive experimental results demonstrate that the proposed CrackCue significantly improves the generalization ability and robustness of the baseline methods. The source code will be publicly available.