Posts by Collection

publications

A frugal approach to music source separation

{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Nathan Souviraà-Labastie", "site"=>"https://fr.linkedin.com/in/nathansouviraa"}, , 2019

During the past years, deep learning brought a big step in performance of music source separation algorithms. A lot has been done on the architecture optimisation, but training data remains an important bias for model comparison. In this work, we choose to work with the frugal and well-known original TasNet neural network and to focus on simple methods to exploit a relatively important dataset. Our results on the MUSDB test set outperform all previous state of the art approaches with extra data on the following source categories: vocals, accompaniment, drums, bass and in average. We believe that our results on how to shape a training set can apply to any type of architecture.

Projection-based classification of surfaces for 3D human mesh sequence retrieval

{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Juan Carlos Alvarez Paiva", "site"=>"https://scholar.google.com/citations?user=OZJvCd8AAAAJ"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, Computer & Graphics, Special Section Shape Modeling International, 2021

We analyze human poses and motion by introducing three sequences of easily calculated surface descriptors that are invariant under reparametrizations and Euclidean transformations. These descriptors are obtained by associating to each finitely-triangulated surface two functions on the unit sphere: for each unit vector u we compute the weighted area of the projection of the surface onto the plane orthogonal to u and the length of its projection onto the line spanned by u. The L2 norms and inner products of the projections of these functions onto the space of spherical harmonics of order k provide us with three sequences of Euclidean and reparametrization invariants of the surface. The use of these invariants reduces the comparison of 3D+time surface representations to the comparison of polygonal curves in Rn. The experimental results on the FAUST and CVSSP3D artificial datasets are promising. Moreover, a slight modification of our method yields good results on the noisy CVSSP3D real dataset.

A Riemannian Framework for Analysis of Human Body Surface

{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}{"name"=>"Alice-Barbara Tumpach", "site"=>"https://math.univ-lille1.fr/~tumpach/Site/home.html"}, IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2022

We propose a novel framework for comparing 3D human shapes under the change of shape and pose. This problem is challenging since 3D human shapes vary significantly across subjects and body postures. We solve this problem by using a Riemannian approach. Our core contribution is the mapping of the human body surface to the space of metrics and normals. We equip this space with a family of Riemannian metrics, called Ebin (or DeWitt) metrics. We treat a human body surface as a point in a “shape space” equipped with a family of Riemmanian metrics. The family of metrics is invariant under rigid motions and reparametrizations; hence it induces a metric on the “shape space” of surfaces. Using the alignment of human bodies with a given template, we show that this family of metrics allows us to distinguish the changes in shape and pose. The proposed framework has several advantages. First, we define a family of metrics with desired invariant properties for the comparison of human shape. Second, we present an efficient framework to compute geodesic paths between human shape given the chosen metric. Third, this framework provides some basic tools for statistical shape analysis of human body surfaces. Finally, we demonstrate the utility of the proposed framework in pose and shape retrieval of human body.

Parameterization Robustness of 3D Auto-Encoders

{"name"=>"Thomas Besnier", "site"=>"https://tbesnier.github.io/"}{"name"=>"Sylvain Arguillère", "site"=>"http://math.univ-lyon1.fr/~arguillere/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, Eurographics Workshop on 3D Object Retrieval (3DOR), 2022

The generation of 3-dimensional geometric objects in the most efficient way is a thriving research topic with, for example, the development of geometric deep learning, extending classical machine learning concepts to non euclidean data such as graphs or meshes. In this short paper, we study the effect of a reparameterization on two popular mesh and point cloud neural networks in an auto-encoder mode: PointNet [QSMG16] and SpiralNet [BBP∗19]. Finally, we tested a modified version of PointNet that takes orientation into account (through coordinates of the normals) as a first step towards the construction of a geometric deep learning model built with a more flexible metric regarding the parameterization. The experimental results on standardized face datasets show that SpiralNet is more robust to the reparametrization than PointNet in this specific context with the proposed reparameterization.

3D Shape Sequence of Human Comparison and Classification Using Current and Varifolds

{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}{"name"=>"Sylvain Arguillère", "site"=>"http://math.univ-lyon1.fr/~arguillere/"}, European Conference on Computer Vision (ECCV) , 2022

Emery Pierson, Mohamed Daoudi, Sylvain Arguillere

Toward Mesh-Invariant 3D Generative Deep Learning with Geometric Measures

{"name"=>"Thomas Besnier", "site"=>"https://tbesnier.github.io/"}{"name"=>"Sylvain Arguillère", "site"=>"http://math.univ-lyon1.fr/~arguillere/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, Computer & Graphics, Special Section 3D Object Retrieval (3DOR), 2023

3D generative modeling is accelerating as the technology allowing the capture of geometric data is developing. However, the acquired data is often inconsistent, resulting in unregistered meshes or point clouds. Many generative learning algorithms require correspondence between each point when comparing the predicted shape and the target shape. We propose an architecture able to cope with different parameterizations, even during the training phase. In particular, our loss function is built upon a kernel-based metric over a representation of meshes using geometric measures such as currents and varifolds. The latter allows to implement an efficient dissimilarity measure with many desirable properties such as robustness to resampling of the mesh or point cloud. We demonstrate the efficiency and resilience of our model with a generative learning task of human faces.

VariGrad: A Novel Feature Vector Architecture for Geometric Deep Learning on Unregistered Data

{"name"=>"Emmanuel Hartmann", "site"=>"https://emmanuel-hartman.github.io/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}, Eurographics Workshop on 3D Object Retrieval (3DOR), 2023

We present a novel geometric deep learning layer that leverages the varifold gradient (VariGrad) to compute feature vector representations of 3D geometric data. These feature vectors can be used in a variety of downstream learning tasks such as classification, registration, and shape reconstruction. Our model’s use of parameterization independent varifold representations of geometric data allows our model to be both trained and tested on data independent of the given sampling or parameterization. We demonstrate the efficiency, generalizability, and robustness to resampling demonstrated by the proposed VariGrad layer.

BaRe-ESA: A Riemannian Framework for Unregistered Human Body Shapes

{"name"=>"Emmanuel Hartmann", "site"=>"https://emmanuel-hartman.github.io/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Martin Bauer", "site"=>"https://www.math.fsu.edu/~bauer/"}{"name"=>"Nicolas Charon", "site"=>"https://www.math.uh.edu/~ncharon/index.html"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, International Conference on Computer Vision (ICCV), 2023

Emmanuel Hartman, Emery Pierson, Martin Bauer, Nicolas Charon, Mohamed Daoudi

Basis restricted elastic shape analysis on the space of unregistered surfaces

{"name"=>"Emmanuel Hartmann", "site"=>"https://emmanuel-hartman.github.io/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Martin Bauer", "site"=>"https://www.math.fsu.edu/~bauer/"}{"name"=>"Nicolas Charon", "site"=>"https://www.math.uh.edu/~ncharon/index.html"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, International Journal of Computer Vision, 2025

Emmanuel Hartman, Emery Pierson, Martin Bauer, Nicolas Charon, Mohamed Daoudi

Measuring Anxiety Levels with Head Motion Patterns in Severe Depression Population

{"name"=>"Fouad Boutaled", "site"=>nil}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Nicolas Doudeau", "site"=>nil}{"name"=>"Clémence Nineuil", "site"=>"https://pro.univ-lille.fr/clemence-nineuil"}{"name"=>"Ali Amad", "site"=>"https://www.aliamad.com/"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, IEEE International Conference on Automatic Face and Gesture Recognition (FG), 2025

Depression and anxiety are prevalent mental health disorders that frequently cooccur, with anxiety significantly influencing both the manifestation and treatment of depression. An accurate assessment of anxiety levels in individuals with depression is crucial to develop effective and personalized treatment plans. This study proposes a new noninvasive method for quantifying anxiety severity by analyzing head movements – specifically speed, acceleration, and angular displacement – during video-recorded interviews with patients suffering from severe depression. Using data from a new CALYPSO Depression Dataset, we extracted head motion characteristics and applied regression analysis to predict clinically evaluated anxiety levels. Our results demonstrate a high level of precision, achieving a mean absolute error (MAE) of 0.35 in predicting the severity of psychological anxiety based on head movement patterns. This indicates that our approach can enhance the understanding of anxiety’s role in depression and assist psychiatrists in refining treatment strategies for individuals.

Wearable-Derived Behavioral and Physiological Biomarkers for Classifying Unipolar and Bipolar Depression Severity

{"name"=>"Yassine Ouzar", "site"=>"https://yassineouzar.github.io/"}{"name"=>"Clémence Nineuil", "site"=>"https://pro.univ-lille.fr/clemence-nineuil"}{"name"=>"Fouad Boutaled", "site"=>nil}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Ali Amad", "site"=>"https://www.aliamad.com/"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, IEEE International Conference on Automatic Face and Gesture Recognition (FG), 2025

Depression is a complex mental disorder characterized by a diverse range of observable and measurable indicators that go beyond traditional subjective assessments. Recent research has increasingly focused on objective, passive, and continuous monitoring using wearable devices to gain more precise insights into the physiological and behavioral aspects of depression. However, most existing studies primarily distinguish between healthy and depressed individuals, adopting a binary classification that fails to capture the heterogeneity of depressive disorders. In this study, we leverage wearable devices to predict depression subtypes-specifically unipolar and bipolar depression-aiming to identify distinctive biomarkers that could enhance diagnostic precision and support personalized treatment strategies. To this end, we introduce the CALYPSO dataset, designed for non-invasive detection of depression subtypes and symptomatology through physiological and behavioral signals, including blood volume pulse, electrodermal activity, body temperature, and three-axis acceleration. Additionally, we establish a benchmark on the dataset using well-known features and standard machine learning methods. Preliminary results indicate that features related to physical activity, extracted from accelerometer data, are the most effective in distinguishing between unipolar and bipolar depression, achieving an accuracy of 96.77%. Temperature-based features also showed high discriminative power, reaching an accuracy of 93.55%. These findings highlight the potential of physiological and behavioral monitoring for improving the classification of depressive subtypes, paving the way for more tailored clinical interventions.

DiffuMatch: Category-Agnostic Spectral Diffusion Priors for Robust Non-rigid Shape Matching

{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Lei Li", "site"=>"https://craigleili.github.io/"}{"name"=>"Angela Dai", "site"=>"https://www.3dunderstanding.org/team.html"}{"name"=>"Maks Ovsjanikov", "site"=>"http://www.lix.polytechnique.fr/~maks/"}, International Conference on Computer Vision (ICCV), 2025

Emery Pierson, Lei Li, Angela Dai, Maks Ovsjanikov

Peripheral skin temperature differentiates unipolar and bipolar depression

{"name"=>"Ali Amad", "site"=>"https://www.aliamad.com/"}{"name"=>"Yassine Ouzar", "site"=>"https://yassineouzar.github.io/"}{"name"=>"Clémence Nineuil", "site"=>"https://pro.univ-lille.fr/clemence-nineuil"}{"name"=>"Fouad Boutaled", "site"=>nil}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Fabien D'Hondt", "site"=>"https://pro.univ-lille.fr/fabien-dhondt"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, Journal of Affective Disorders, 2025

We read with interest the recent study by Valenzuela-Pascual et al. reporting an increase in peripheral skin temperature during manic episodes compared to depressive episodes in patients with bipolar disorder (Valenzuela-Pascual et al., 2025). Their results suggest that skin temperature, passively recorded by wearable sensors, could serve as a state-dependent physiological marker of manic versus depressive phases. This aligns with recent efforts to identify objective physiological correlates of psychiatric symptoms using unobtrusive, ecological tools. We would like to offer a complementary perspective based on data from the CALYPSO study, in which peripheral skin temperature was used to explore trait-level physiological differences between unipolar and bipolar depression.

PaNDaS: Learnable Deformation Modeling with Localized Control

{"name"=>"Thomas Besnier", "site"=>"https://tbesnier.github.io/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Sylvain Arguillère", "site"=>"http://math.univ-lyon1.fr/~arguillere/"}{"name"=>"Maks Ovsjanikov", "site"=>"https://www.lix.polytechnique.fr/~maks/"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}, IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

Thomas Besnier, Emery Pierson, Sylvain Arguillère, Maks Ovsjanikov, Mohamed Daoudi

Unfolding Generative Flows with Koopman Operators: Fast and Interpretable Sampling

{"name"=>"Erkan Turan", "site"=>"https://ehturan.github.io/"}{"name"=>"Aristotelis Siozopoulos"}{"name"=>"Louis Martinez", "site"=>"https://scholar.google.com/citations?user=UwfztWsAAAAJ"}{"name"=>"Julien Gaubil", "site"=>"https://jgaubil.com/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Maks Ovsjanikov", "site"=>"https://www.lix.polytechnique.fr/~maks/"}, , 2026

Continuous Normalizing Flows (CNFs) enable elegant generative modeling but remain bottlenecked by their iterative nature requiring costly sampling and lacking interpretability of the intermediate states. Recent approaches accelerate sampling by straightening trajectories or distilling endpoints, yet they treat the original generative process as a black box, discarding the teacher’s intermediate dynamics. We propose a fundamentally different perspective: globally linearizing flow dynamics via Koopman theory to achieve trajectory-preserving linearization. By lifting Conditional Flow Matching (CFM) into a higher-dimensional Koopman space, we represent its evolution with a single linear operator. Crucially, unlike boundary-only distillation, our method enforces infinitesimal consistency with the teacher’s vector field along the full generative path. We derive a practical, simulation-free training objective that ensures this global alignment and yields two key benefits. First, sampling becomes one-step and parallelizable. Second, because the linearization is faithful to the dynamics, the Koopman operator provides unique insights on the generation. We demonstrate that this structure enables novel applications unavailable in prior approaches, including discovery of semantically coherent editing directions, inversion with a teacher-aligned linear operator and class-conditional spectral signatures. Empirically, our approach achieves competitive sample quality, while enabling spectral analysis and control of the entire trajectories of generative flows.

PatchAlign3D: Local Feature Alignment for Dense 3D Shape Understanding

{"name"=>"Souhail Hadgi", "site"=>"https://souhail-hadgi.github.io/"}{"name"=>"Bingchen Gong", "site"=>"https://s2.hk/"}{"name"=>"Ramana Sundararaman", "site"=>"https://sentient07.github.io/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Lei Li", "site"=>"https://craigleili.github.io/"}{"name"=>"Peter Wonka", "site"=>"https://peterwonka.net/"}{"name"=>"Maks Ovsjanikov", "site"=>"https://www.lix.polytechnique.fr/~maks/"}, IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

Q: What is the missing piece in current 3D foundation models?

Beyond ReLU: Bifurcation, Oversmoothing, and Topological Priors

{"name"=>"Erkan Turan", "site"=>"https://ehturan.github.io/"}{"name"=>"Gaspard Abel", "site"=>"https://www.gas-abel.fr/"}{"name"=>"Maysam Behmanesh", "site"=>"https://maysambehmanesh.github.io/"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Maks Ovsjanikov", "site"=>"https://www.lix.polytechnique.fr/~maks/"}, , 2026

Graph Neural Networks (GNNs) learn node representations through iterative network-based message-passing. While powerful, deep GNNs suffer from oversmoothing, where node features converge to a homogeneous, non-informative state. We re-frame this problem of representational collapse from a bifurcation theory perspective, characterizing oversmoothing as convergence to a stable ``homogeneous fixed point.’’ Our central contribution is the theoretical discovery that this undesired stability can be broken by replacing standard monotone activations (e.g., ReLU) with a class of functions. Using Lyapunov-Schmidt reduction, we analytically prove that this substitution induces a bifurcation that destabilizes the homogeneous state and creates a new pair of stable, non-homogeneous patterns that provably resist oversmoothing. Our theory predicts a precise, nontrivial scaling law for the amplitude of these emergent patterns, which we quantitatively validate in experiments. Finally, we demonstrate the practical utility of our theory by deriving a closed-form, bifurcation-aware initialization and showing its utility in real benchmark experiments.

A Non-Invasive 3D Gait Analysis Framework for Quantifying Psychomotor Retardation in Major Depressive Disorder

{"name"=>"Fouad Boutaleb"}{"name"=>"Emery Pierson", "site"=>"https://daidedou.github.io/"}{"name"=>"Mohamed Daoudi", "site"=>"https://sites.google.com/view/mohameddaoudi"}{"name"=>"Clémence Nineuil", "site"=>"https://pro.univ-lille.fr/clemence-nineuil"}{"name"=>"Ali Amad", "site"=>"https://www.aliamad.com/"}{"name"=>"Fabien D'Hondt", "site"=>"https://pro.univ-lille.fr/fabien-dhondt"}, , 2026

Predicting the status of Major Depressive Disorder (MDD) from objective, non-invasive methods is an active research field. Yet, extracting automatically objective, interpretable features for a detailed analysis of the patient state remains largely unexplored. Among MDD’s symptoms, Psychomotor retardation (PMR) is a core item, yet its clinical assessment remains largely subjective. While 3D motion capture offers an objective alternative, its reliance on specialized hardware often precludes routine clinical use. In this paper, we propose a non-invasive computational framework that transforms monocular RGB video into clinically relevant 3D gait kinematics. Our pipeline uses Gravity-View Coordinates along with a novel trajectory-correction algorithm that leverages the closed-loop topology of our adapted Timed Up and Go (TUG) protocol to mitigate monocular depth errors. This novel pipeline enables the extraction of 297 explicit gait biomechanical biomarkers from a single camera capture. To address the challenges of small clinical datasets, we introduce a stability-based machine learning framework that identifies robust motor signatures while preventing overfitting. Validated on the CALYPSO dataset, our method achieves an 83.3% accuracy in detecting PMR and explains 64% of the variance in overall depression severity (R^2=0.64). Notably, our study reveals a strong link between reduced ankle propulsion and restricted pelvic mobility to the depressive motor phenotype. These results demonstrate that physical movement serves as a robust proxy for the cognitive state, offering a transparent and scalable tool for the objective monitoring of depression in standard clinical environments.

teaching

Introduction to basic mathematics

Undergraduate course, Université de Lille, 2021

Teaching of first years bachelors, and assistance for exercises. Design of midterm exams.

Computer Science Teaching assistant

Undergraduate course, Lille University Technological Institute (IUT de Lille) - Computer Science Department, 2022

Teaching assistant during lab sessions: Introduction to computer sience, object oriented programming, web design. Java, HTML, CSS.

Geometric deep learning Teaching assistant

Master course, Maths, Vision Learning master (MVA), 2024

Teaching assistant during lab sessions: Introduction to geometric objects, learning on shapes, neural shape representations. Website