Santiago VELASCO-FORERO
 |
Ph.D. / HDR |
About me:
Born in Colombia, I am currently Fellow Researcher (Chargé de Recherche in the French system) at Mines Paris PSL Research University. Before coming to Mines Paris, I have worked as a postdoc researcher at ITWM, which is part of the Fraunhofer Institute in Germany, and at the Mathematical Department of the National University of Singapore (NUS). In my research, I am interested in understanding and designing methods to solve problems on images, 3D cloud points and graphs. I have a preference for the method based on interactions between nonlinear operators with high geometric interpretation and those based on deep learning.
Interested in working with me?
To apply to work with me as a PhD student, email me your CV (pdf), and a statement of research interests, indicating if you are interested (or not) in doing research in the framework of a CIFRE thesis.
HDR (Habilitation à diriger des recherches)
Institut Polytechnique de Paris (26-09-2022)
Contributions to Mathematical Morphology
Jury:
Isabelle BLOCH, Professeur, Sorbonne Université, Paris
Pierre SOILLE, Project Leader, HDR, Joint Research Centre of the European Commission
Yann GOUSSEAU, Professeur, Télécom Paris
Nicolas PASSAT, Professeur, Université de Reims Champagne-Ardenne
Jean-Michel MOREL, Professeur, École Normale Supérieure
Stephane MALLAT, Professeur, Collège de France
Publications
Next events
(02-10-2024) Organizer of the Workshop “Symmetry and equivariance for Deep Learning”, as part of CaLISTA Workshop 2024 “Geometry and Machine Learning”.
News
(19-03-2024) Invited plenary talk in the 37ème Congrès Français sur les Aérosols (in French), Paris, France.
(02-02-2024) Counting melanocytes with trainable h-maxima and connected components counting layers (DGMM2024) Oral
(02-02-2024) Group Equivariant Networks Using Morphological Operators (DGMM2024) Oral
(02-02-2024) Spatial Reasoning Loss for Weakly Supervised Segmentation of Skin Histological Images (ISBI 2024) Oral
(02-02-2024) Choquet Capacity networks for random point process classification (CMDS-14)
(01-01-2024) DEEPORDER: A collaborative project between Unicampinas and Mines has been selected for ANR 2023 funding.
(01-12-2023) Roto-translation Equivariant YOLO for Aerial Images (Extended abstract on NIPS-workshop)
(06-10-2023) Invited talk to the XI Encuentro Nacional de Matemáticas y Estadística (in Spanish), Ibagué, Colombia
(03-10-2023) Co-organizing workshop at ICCV 2023 https:www.latinxinai.org/iccv-2023
(01-09-2023) Rotation-invariant Hierarchical Segmentation on Poincaré Ball for 3D Point Cloud, ICCV 2023 SHARP Workshop (Oral presentation) Code
(30-08-2023) Can generalized divergences help for invariant neural networks? GSI 2023. (Oral presentation) Code
(20-06-2023) Invited Plenary talk at the GDMM 2023 (in French)
(22-05-2023) GenHarris-ResNet: A Rotation Invariant Neural Network Based on Elementary Symmetric Polynomials SSVM 2023 (Italy). (Oral presentation) Code
(10-05-2023) Invited to talk in the JPAL Seminar about Geometry-Informed and Physics-Informed Neural Networks in Palaiseau (France).
(21-04-2023) Participated in the panel “Carving out a niche as a researcher” of AISTATS2023 in Valencia (Spain).
(21-11-2022) Moving Frame Net: SE (3)-Equivariant Network for Volumes has been accepted in the NeurIPS 2022 Workshop.
(21-11-2022) Scale Equivariant U-Net has been accepted in the BMVC2022.
(21-11-2022) Fixed Point Layers for Geodesic Morphological Operations has been accepted in the BMVC2022.
(5-10-2022) Instance segmentation of 3D woven fabric from tomography images by Deep Learning and morphological pseudo-labeling accepted to Composites Part B: Engineering journal.
(24-10-2022) MorphoActivation: Generalizing ReLU activation function by mathematical morphology has been accepted in the DGMM2022.
(24-10-2022) Morphological adjunctions represented by matrices in max-plus algebra for signal and image processing has been accepted in the DGMM2022.
(16-10-2022) Differential invariants for SE (2)-equivariant networks has been accepted in IEEE ICIP 2022 (Oral)
(16-10-2022) Fully Trainable Gaussian Derivative Convolutional Layer has been accepted in IEEE ICIP 2022
(28-09-2022) I have defended my HDR!
(19-09-2022) ECML PKDD 2022 Near Out-of-distribution detection for low-resolution radar micro-doppler signatures (Oral)
(01-04-2022) DarthBoard based ground detection on 3D Point Cloud has been accepted in the ISPRS 2022.
(01-01-2022) Yuriy Sinchuk joins our research group in a Postdoctoral position on Deep Learning for Material Science.
(01-01-2022) ICLR 2021 Challenge for Computational Geometry & Topology: Design and Results
(01-12-2021) Paris-CARLA-3D: A Real and Synthetic Outdoor Point Cloud Dataset for Challenging Tasks in 3D Mapping
(28-10-2021) Ph.D. thesis defense of David Duque -> 3D urban scene understanding by analysis of LiDAR, color and hyperspectral data.
(25-09-2021) Course on Anomaly Detection in the Colombian Symposium on Statistics 2021.
(25-06-2021) Journal Paper on Learnable Empirical Mode Decomposition based on Mathematical Morphology accepted for publication in the SIAM Journal on Imaging Sciences.
(01-05-2021) Ayoub Rhim from Ecole des Pons ParisTech is now a master intern in a PGMO Projet in collaboration with J. Angulo
(01-04-2021) Josselin Lefevre from ESSIE Paris is now a master intern in collaboration with S. Blusseau and Safran Aircraft Engines.
(23-03-2021) Ph.D. Thesis defense of Leonardo Gigli -> Contributions to graph-based hierarchical analysis for images and 3D point clouds
(05-03-2021) Two accepted conferences paper in the DGMM 2021
(01-03-2021) Accepted paper on the 2021 IEEE ICASSP. NNAKF: A neural network adapted Kalman filter for target tracking.
(01-03-2021) A new conference publication on modified Jaccard loss for semantic segmentation in VISAPP 2021. PDF
(29-07-2020) SHREC 2020 Track: 3D Point Cloud Semantic Segmentation for Street Scenes has been accepted to Computers and Graphics Journal
(29-07-2020) SHREC’20 track: Retrieval of digital surfaces with similar geometric reliefs has been accepted to Computers and Graphics Journal
(27-07-2020) Approximating morphological operators with part-based representations learned by asymmetric auto-encoders has been accepted for publication in Mathematical Morphology: Theory and Applications.
(01-07-2020) On minimum spanning tree streaming for hierarchical segmentation has been accepted in Pattern Recognition Letters.
(01-06-2020) From unsupervised to semi-supervised anomaly detection methods for High Resolution Range Profiles has been accepted on the IEEE Radar Conference 2020
(01-05-2020) Road segmentation on low resolution lidar point clouds for autonomous vehicles is an accepted paper on the ISPRS2020
(01-05-2020) Romain Pagès from École Centrale de Lyon is intern in a PGMO Projet in collaboration with J. Angulo
(01-01-2020) Journal paper on Morphological Hierarchical Image characterization has been published in Pattern Recognition Letters
(30-10-2019) Journal paper on Prior-based Hierarchical Segmentation in Mathematical Morphology: Theory and Applications
(10-07-2019) The first results in D.Duque's thesis will be presented at the ICIAP Conference.
(10-07-2019) Award: Co-author of the ISMM 2019 Best Student Paper Award to Bastien Ponchon for Part-based approximations for morphological operators using asymmetric auto-encoders.
(05-06-2019) A journal paper has been accepted for publication in the Astronomy & Astrophysics journal.
(05-06-2019) Invited to talk at the internal seminar in the CMLA ENS Cachan, France.
(01-06-2019) Invited Oral talk at the ICSIA-2019 in Aarhus, Denmark.
(15-03-2019) Two accepted conference papers in the ISMM 2019.
(15-03-2019) Zyad Haddad from ESIEE is an intern in collaboration with S. Blusseau on Deep Learning for Material Science.
(15-02-2019) Tristan Lazard from ENS/University PSL is an intern in collaboration with S. Blusseau, and E. Decenciere.
(12-01-2019) A collaborative work with L'Oréal R&I has been accepted in IEEE ISBI 2019.
(06-11-2018) Invited plenary talk at Colombian First Meeting on Applied Mathematics
(10-10-2018) A journal paper has been accepted for publication in the Journal of Sensors. This is part of a collaborative work with Rioja University.
(06-10-2018) Three accepted papers in the IEEE International Conference on Image Processing 2018
(06-10-2018) A collaborative work with L'Oréal R&I has been accepted in the ACIVS 2018 conference.
(26-07-2018) Invited professor in the National University of Colombia (Department of Statistics).
(25-05-2018) (14h30 L213 60 Boulevard Saint-Michel, 75006, Paris) A. Fehri's thesis defense.
(01-05-2018) Bastien Ponchon from ENS Paris-Saclay is an intern in a IMT Project in collaboration with I. Bloch, S. Blusseau, and J. Angulo.
(01-04-2018) Ulrike Kuchner from University of Vienna is now a postdoc researcher at CMM working on applications of Deep Learning for Astronomy.
(01-04-2018) Yunxiang Zhang from Ecole Polytechnique is an intern in a IMT Project in collaboration with I. Bloch, S. Blusseau, and J. Angulo.
(01-04-2018) Mael Abgrall is an intern at CMM/MINES ParisTech in collaboration with E. Decenciere.
(01-03-2018) Award: Outstanding reviewer in Expert Systems with Applications
(01-02-2018) Sami Jouaber from MINES ParisTech is an intern in a PGMO Project in collaboration with S. Bonnabel and J. Angulo.
(01-01-2018) Award: Outstanding reviewer in Pattern Recognition
(01-08-2017) Luis Javier Morantes from ITM-Medellin and University of Antioquia is visiting our lab.
(01-03-2017) Umang Aggarwal is an intern at Mines ParisTech/Ecole Normale Supérieure collaborative project between CMM and EREEP.
(01-02-2017) Samy Blusseau is now a postdoc researcher in CMM/Telecom ParisTech collaborative project with J. Angulo (CMM), I.Bloch and Yann Gousseau (Telecom Paristech).
(01-07-2016) Program Co-Chair in the 13th International Symposium on Mathematical Morphology
(15-01-2016) Borja Millan-Prior from University of La Rioja is visiting our lab.(Project information: http://televitis.unirioja.es/en/)
Contact
santiago dot velasco at mines-paristech dot fr
Office telephone: (+33) 1 64 69 47 06
Office: B-205 (Directions)
Research interests
I am interested in collaborating with mathematicians, statisticians, scientists and engineers on diverse projects including:
Image Processing
Deep Learning
Multivariate Statistics
Graph Signal Processing
Artificial Intelligence
Teaching
(2020-2021-2022-2023-2024) Deep Learning for Image Analysis (MINES/ Master IASD - LAMSADE)
(2018-2019-2022-2023-2024) Deep Learning for Image Analysis (MINES / ATHENS )
(2018) Introduction of Big Data and Applications
Ph.D. Students
Valentin Penaud Polge (Coadvisor with J. Angulo) (2021-2023)
Joao Casagrande Bertoldo (Coadvisor with E. Decencière and J. Angulo) (2022-2024)
Stéfan Baudier (Coadvisor with J. Angulo) (2022-2024)
Edwyn Brient(2024-2027)
Pierre Onghena (Coadvisor with B. Marcotegui) (2024-2027)
Arthur Roblin ( Coadvisor with G. Dougniaux et J. Baccou) (2024-2027)
Ph.D. Alumni
(18-01-2023) Martin Bauw (Coadvisor J.Angulo )
Thesis title: “One-class classification for low resolution targets discrimination with limited supervision in pulse Doppler radars”
Jury: Marius Kloft, TU Kaiserslautern, Germany
Jean-Philippe Ovarlez, ONERA, France
Florence Tupin, Télécom Paris, LTCI, Institut Polytechnique de Paris, France
Claude ADNET, Thales LAS, France
(13-12-2022) Mateus Sangalli (Coadvisor J.Angulo and S. Blusseau)
Thesis title: “Equivariant Deep Learning Based on Scale-Spaces and Moving Frames: Applications to Image Processing”
Jury: Petros Maragos - University of Athens, Greece
Tony Lindeberg - KTH Royal Institute of Technology, Sweden
Nina Miolane - UC Santa Barbara, USA
Marcos Eduardo Valle - University of Campinas, Brasil
(20-09-2022) Sami Jouaber (Coadvisor S.Bonnabel)
Thesis title: “Utilisation de méthodes d'apprentissage automatique pour des tâches de pistage radar et de classification d'aéronefs”
Jury: Audrey GIREMUS - Université de Bordeaux, IMS
Frédéric BOUCHARA - Université de Toulon, LIS
François SEPTIER Université Bretagne Sud, LMBA
Marion PILTE - THALES LAS
Frédéric LIVERNET - DGA
(28-10-2021) David Duque (Coadvisor with B. Marcotegui)
Thesis title: “3D urban scene understanding by analysis of LiDAR, color and hyperspectral data”.
Jury: A. Nuchter ( University of Würzburg, Germany),
S. Lefèvre (Université Bretagne Sud),
Pascal MONASSE(Ecole Nationale des Ponts et Chaussées)
Paul CHECCHIN (Université Clermont Auvergne)
Jean-Emmanuel DESCHAUD(Mines ParisTech)
B. Marcotegui (Mines ParisTech)
(23-03-2021) Leonardo Gigli (Coadvisor with B. Marcotegui)
Thesis title: “Contributions to graph-based hierarchical analysis for images and 3D point clouds”.
Jury: A. Plaza ( University of Extremadura),
P. Soille (JRC),
S. Lefèvre (Université Bretagne Sud),
B. Perret (ESIEE Paris),
Ravi Kiran (NavyaTech)
J. Angulo (Mines ParisTech)
Funding
M3S Project (2015-2016) Molecular detection with Multimodal microscopy scanner (http://www.m3scan.eu/home)
COMSYS Project (2015-2016) (Association Institute Carnot M.I.N.E.S)
REPLICA Project (2019-2021) (https://replica-project.com/)
DEEPORDER Project (2024-2028) (DeepOrder.html)
Professional service:
Reviewer for Image Processing and Signal Processing journals: Journal of Mathematical Imaging and Vision IEEE-Transactions on Image Processing IEEE-Selected Topics in Signal Processing IEEE-Transactions on Signal Processing IEEE Signal Processing Letters Digital Signal Processing Image Analysis and Stereology Information Fussion
Remote Sensing journals: IEEE-Transactions on Geoscience and Remote Sensing IEEE- Special Topics in Remote Sensing International Journal of Remote Sensing
Pattern Recognition and other topics journals: IEEE Transactions on Pattern Analysis and Machine Intelligence IEEE Transactions on Neural Netw. and Learning Systems Neurocomputing Pattern Recognition Pattern Recognition Letters Information Sciences Applied Mathematical Modelling