Publications | Stéphane Herbin

Smoothing Continual Segmentation Oscillations with Latent Domain PPCA Decoder

We study Domain Incremental Learning for the semantic segmentation of Earth Observation images. We demonstrate that controlling the …

Marie-Ange Boum, Pierre Fournier, Dawa Derksen, Stéphane Herbin

Measuring and addressing information leakage in concept bottleneck models

Raffael Schoen, Baptiste Abeloos, Stéphane Herbin

Bridging the Modality Gap: Training-free Adaptation of Vision-Language Models for Remote Sensing via Visual Prototypes

Clément Barbier, Baptiste Abeloos, Stéphane Herbin

Interpretable Image Classification through an Argumentative Dialog between Encoders

We address the problem of designing interpretable algorithms for image classification. Modern computer vision algorithms implement …

Dao Thauvin, Stéphane Herbin, Wassila Ouerdane, Céline Hudelot

How to Efficiently Adapt Foundation Models to Remote Sensing for Object Detection?

Vision-Language Models (VLMs) trained on millions of images excel at generalization in various domains. However, a significant …

Clément Barbier, Baptiste Abeloos, Stéphane Herbin

Continual Learning in Remote Sensing : Leveraging Foundation Models and Generative Classifiers to Mitigate Forgetting

Continual learning in dynamic environments is a challenge for large-scale machine learning models. This research addresses Domain …

Marie-Ange Boum, Stéphane Herbin, Pierre Fournier, Pierre Lassalle

Thinking the Certification Process of Embedded ML-based Aeronautical Components Using AIDGE, a French Open and Sovereign AI Platform

AIDGE is a novel software development platform for embedded Artificial Intelligence (AI). It is designed to import or even learn Deep …

Filipo Studzinski Perotto, Anthony Fernandes Pires, Jean-Loup Farges, Youcef Bouchebaba, Mohammed Belcaid, Eric Bonnafous, Claire Pagetti, Frédéric Boniol, Xavier Pucel, Adrien Chan-Hon-Tong, Stéphane Herbin, Mario Cassaro, Sofiane Kraiem

Optimal Transport Based Adversarial Patch to Leverage Large Scale Attack Transferability

Adversarial patch attacks, where a small patch is placed in the scene to fool neural networks, have been studied for numerous …

Pol Labarbarie, Adrien Chan-Hon-Tong, Stéphane Herbin, Milad Leyli-Abadi

Efficient Exploration of Image Classifier Failures with Bayesian Optimization and Text-to-Image Models

Adrien Le Coz, Houssem Ouertatani, Stéphane Herbin, Faouzi Adjed

Trusting Machine-Learning Applications in Aeronautics

A general recommendation from the French office for aeronautical and space standardization (BNAE) is being drawn up by experts from …

Karim Benmeziane, Patrick Fabiani, Stéphane Herbin, Jérôme Lacaille, Emmanuel Ledinot

Knowledge Informed Sequential Scene Graph Verification Using VQA

We propose a new task, non localized scene graph verification, whose objective is to provide a justified expression of inconsistencies …

Dao Thauvin, Stéphane Herbin

Explaining an Image Classifier with a Generative Model Conditioned by Uncertainty

We propose to condition a generative model by a given image classifier uncertainty in order to analyze and explain its behavior. …

Adrien Le Coz, Stéphane Herbin, Faouzi Adjed

Leveraging Generative Models to Characterize the Failure Conditions of Image Classifiers

We address in this work the question of identifying the failure conditions of a given image classifier. To do so, we exploit the …

Adrien Le Coz, Stéphane Herbin, Faouzi Adjed

Explaining Object Detectors: The Case of Transformer Architectures

Object detection is a complex visual function that has applications in many safety-critical domains such as autonomous driving or …

Baptiste Abeloos, Stéphane Herbin

Benchmarking and Deeper Analysis of Adversarial Patch Attack on Object Detectors

Adversarial attacks (either norm bounded or patch-based) have received much attention from the computer vision community over the last …

Pol Labarbarie, Adrien Chan-Hon-Tong, Stéphane Herbin, Milad Leyli-Abadi

Semi-Supervised Class Incremental Learning

This paper makes a contribution to the problem of incremental class learning, the principle of which is to sequentially introduce …

Alexis Lechat, Stéphane Herbin, Frédéric Jurie

Pseudo-Labeling for Class Incremental Learning

Class Incremental Learning (CIL) consists in training a model iteratively with limited amount of data from few classes that will never …

Alexis Lechat, Stéphane Herbin, Frédéric Jurie

Towards Autonomous PErceptual Systems

The objective of this document is to introduce the principle of Autonomous PErceptual System (APES) as an object of study. The …

Stéphane Herbin

Challenges in the Certification of Computer Vision-Based Systems for Civil Aeronautics

F. Boniol, A. Chan-Hon-Tong, A. Eudes, S. Herbin, G. Le Besnerais, C. Pagetti, M. Sanfourche

Artificial Intelligence and Decision Making

Stéphane Herbin, Jean-Loup Farges

Method for Detecting Cells Having at Least One Anomaly in a Cytological Sample

Zehira HADDAD, Stephane Herbin, Adrien Chan-Hon-Tong

Adaptation Du Problème de Questions-Réponses Visuelles à Un Contexte d'apprentissage Continu

Le problème de questions-réponses visuelles (QRV) constitue un nouveau défi en intelligence artificielle. Il requiert du système des …

Alexis Lechat, Stéphane Herbin, Frédéric Jurie

Zero-Shot Classification by Generating Artificial Visual Features

Maxime Bucher, Stéphane Herbin, Frédéric Jurie

Semantic Bottleneck for Computer Vision Tasks

This paper introduces a novel method for the representation of images that is semantic by nature, addressing the question of …

Maxime Bucher, Stéphane Herbin, Frédéric Jurie

On-Line Fusion of Trackers for Single-Object Tracking

Visual object tracking is a fundamental function of computer vision that has been the object of numerous studies. The diversity of the …

Isabelle Leang, Stéphane Herbin, Benoît Girard, Jacques Droulez

Practical Scheduling of Computer Vision Functions

Adrien Chan-Hon-Tong, Stéphane Herbin

Generating Visual Representations for Zero-Shot Classification

This paper addresses the task of learning an image classifier when some categories are defined by semantic descriptions only (e.g. …

Maxime Bucher, Stéphane Herbin, Frédéric Jurie

Processing of Extremely High-Resolution LiDAR and RGB Data: Outcome of the 2015 IEEE GRSS Data Fusion Contest--Part A: 2-D Contest

M. Campos-Taberner, A. Romero-Soriano, C. Gatta, G. Camps-Valls, A. Lagrange, B. Le Saux, A. Beaupère, A. Boulch, A. Chan-Hon-Tong, S. Herbin, H. Randrianarivo, M. Ferecatu, M. Shimoni, G. Moser, D. Tuia

Improving Semantic Embedding Consistency by Metric Learning for Zero-Shot Classiffication

This paper addresses the task of zero-shot image classification. The key contribution of the proposed approach is to control the …

Maxime Bucher, Stéphane Herbin, Frédéric Jurie

Hard Negative Mining for Metric Learning Based Zero-Shot Classification

Zero-Shot learning has been shown to be an efficient strategy for domain adaptation. In this context, this paper builds on the recent …

Maxime Bucher, Stéphane Herbin, Frédéric Jurie

Tracking Based Sparse Box Proposal for Time Constraint Detection in Video Stream

Search and Rescue or surveillance applications from en embedded moving camera yield challenging computer vision problems as both very …

Adrien Chan-Hon-Tong, Stéphane Herbin

Robust Fusion of Trackers Using Online Drift Prediction

Visual object tracking is a standard function of computer vision that has been the source of numerous propositions. This diversity of …

Isabelle Leang, Stéphane Herbin, Benoît Girard, Jacques Droulez

Exemplar Based Metric Learning for Robust Visual Localization

This paper presents an exemplar based metric learning framework dedicated to robust visual localization in complex scenes, e.g. street …

Cedric Le Barz, Nicolas Thome, Matthieu Cord, Stéphane Herbin, Martial Sanfourche

Benchmarking Classification of Earth-Observation Data: From Learning Explicit Features to Convolutional Networks

Adrien Lagrange, Bertrand Le Saux, Anne Beaupere, Alexandre Boulch, Adrien Chan-Hon-Tong, Stéphane Herbin, Hicham Randrianarivo, Marin Ferecatu

Absolute Geo-Localization Thanks to Hidden Markov Model and Exemplar-Based Metric Learning

This paper addresses the problem of absolute visual ego-localization of an autonomous vehicle equipped with a monocular camera that has …

Cedric Le Barz, Nicolas Thome, Matthieu Cord, Stéphane Herbin, Martial Sanfourche

Global Robot Ego-localization Combining Image Retrieval and HMM-based Filtering

This paper addresses the problem of global visual ego-localization of a robot equipped with a monocular camera that has to navigate …

Cédric Le Barz, Nicolas Thome, Matthieu Cord, Stéphane Herbin, Martial Sanfourche

Fine-Grained Recognition by Sequential Hypothesis Rejection and Foveated Vision on Parts

Visual foveation is the association of a spatially variant resolution sensor – the retina – and a dynamic controlled …

Stéphane Herbin

Vehicle Recognition in Video Surveillance

Stéphane Herbin

Robust Abandoned Object Detection Integrating Wide Area Visual Surveillance and Social Context

This paper presents a video surveillance framework that robustly and efficiently detects abandoned objects in surveillance scenes. The …

James M. Ferryman, David C. Hogg, Jan Sochman, Ardhendu Behera, José A. Rodríguez-Serrano, Simon F. Worgan, Longzhen Li, Valerie Leung, Murray Evans, Philippe Cornic, Stéphane Herbin, Stefan Schlenger, Michael Dose

Semantic Hierarchies for Image Annotation: A Survey

In this survey, we argue that using structured vocabularies is capital to the success of image annotation. We analyze literature on …

Anne-Marie Tousch, Stéphane Herbin, Jean-Yves Audibert

Scene Understanding from Aerospace Sensors: What Can Be Expected?

S. Herbin, F. Champagnat, J. Israel, F. Janez, B. Le Saux, V. Leung, A. Michel

Multimodalités et Ontologies

J. Besombes, M. Campedel, V. Dragos, S. Herbin, O. Poirel, A.-M. Tousch

Flexible Tracklet Association for Complex Scenarios Using a Markov Logic Network

Multi-object tracking, despite significant advances in recent years, is still an area where improvements can be made. The additional …

Valerie Leung, Stéphane Herbin

Context-Driven Moving Object Detection in Aerial Scenes with User Input

Aerial video sequences are a common source for applications such as intelligence, surveillance or search and rescue. Their off-line …