Pablo Mesejo Santiago
Présentation
My name is Pablo Mesejo Santiago and I currently hold a Marie Curie Experienced Researcher position at the University of Granada (Spain), one of the top institutions in computer science and engineering. My principal research areas of interest are computer vision, machine learning and computational intelligence methods applied to image analysis problems (mainly in the biomedical domain). Typical tools I use in my research are stochastic optimization algorithms, deep and shallow neural networks, and ensemble classifiers. During my career I have tackled numerous challenging problems, e.g. the automatic segmentation of anatomical structures in biomedical images (PhD at University of Parma, performed as a Marie Curie Early Stage Researcher, 2010-13), the classification of gastrointestinal lesions from endoscopic videos (postdoc at University of Auvergne Clermont-Ferrand I, 2013-14), the estimation of biophysical parameters from fMRI signals (postdoc at Inria, 2014-16), and the integration of deep learning into probabilistic generative models for visual and audio recognition in human-robot interaction (starting researcher position at Inria, 2016-18). More information about me and my publications can be found in the following links: Google Scholar, ORCID, Linkedin, DBLP, ResearchGate and ResearcherID.
Publications
Publications
|
|
Learning to Reason with Transformers via Search Inductive Biases: A ProposalAAAI25 Workshop on Planning in the Era of LLMs (LM4Plan), Association for the Advancement of Artificial Intelligence (AAAI), Feb 2025, Philadelphia (Pennsylvania), United States |
|
|
Learning Visual Voice Activity Detection with an Automatically Annotated DatasetICPR 2020 - 25th International Conference on Pattern Recognition, Jan 2021, Milano, Italy. pp.4851-4856, ⟨10.1109/ICPR48806.2021.9412884⟩ |
|
|
Understanding Priors in Bayesian Neural Networks at the Unit LevelICML 2019 - 36th International Conference on Machine Learning, Jun 2019, Long Beach, United States. pp.6458-6467 |
|
|
Extended Gaze Following: Detecting Objects in Videos Beyond the Camera Field of ViewFG 2019 - 14th IEEE International Conference on Automatic Face and Gesture Recognition, May 2019, Lille, France. pp.1-8, ⟨10.1109/FG.2019.8756555⟩ |
|
|
Deep Reinforcement Learning for Audio-Visual Gaze ControlIROS 2018 - IEEE/RSJ International Conference on Intelligent Robots and Systems, Oct 2018, Madrid, Spain. pp.1555-1562, ⟨10.1109/IROS.2018.8594327⟩ |
|
|
DeepGUM: Learning Deep Robust Regression with a Gaussian-Uniform Mixture ModelECCV 2018 - European Conference on Computer Vision, Sep 2018, Munich, Germany. pp.205-221, ⟨10.1007/978-3-030-01228-1_13⟩ |
|
|
Bayesian neural networks become heavier-tailed with depthNeurIPS 2018 - Thirty-second Conference on Neural Information Processing Systems, Dec 2018, Montréal, Canada. pp.1-7 |
|
|
Bayesian neural network priors at the level of unitsAABI 2018 - 1st Symposium on Advances in Approximate Bayesian Inference, Dec 2018, Montréal, Canada. pp.1-6 |
|
|
Deep Mixture of Linear Inverse Regressions Applied to Head-Pose EstimationIEEE Conference on Computer Vision and Pattern Recognition, Jul 2017, Honolulu, Hawaii, United States. pp.7149-7157, ⟨10.1109/CVPR.2017.756⟩ |
|
|
Estimating Biophysical Parameters from BOLD Signals through Evolutionary-Based Optimization18th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI’15), Oct 2015, Munich, Germany. pp.528-535, ⟨10.1007/978-3-319-24571-3_63⟩ |
|
|
Automatic evolutionary medical image segmentation using deformable models16th IEEE Congress on Evolutionary Computation (CEC’14), Jul 2014, Beijing, China. pp.97-104, ⟨10.1109/CEC.2014.6900466⟩ |
|
|
An Analysis of Errors in Graph-Based Keypoint Matching and Proposed Solutions13th European Conference on Computer Vision (ECCV’14), Sep 2014, Zurich, Switzerland. pp.138-153, ⟨10.1007/978-3-319-10584-0_10⟩ |
|
|
Segmentation of histological images using a metaheuristic-based level set approach15th Genetic and Evolutionary Computation Conference companion (GECCO’13), Jul 2013, Amsterdam, Netherlands. pp.1455-1462, ⟨10.1145/2464576.2466808⟩ |
|
|
Algorithm configuration using GPU-based metaheuristics15th Genetic and Evolutionary Computation Conference companion (GECCO’13), Jul 2013, Amsterdam, Netherlands. pp.221-222, ⟨10.1145/2464576.2464682⟩ |
|
|
An experimental study on the automatic segmentation of in situ hybridization-derived images1st International Conference on Medical Imaging using Bio-Inspired and Soft Computing (MIBISOC’13), May 2013, Brussels, Belgium. pp.153-160 |
|
|
GPU-Based Automatic Configuration of Differential Evolution: A Case Study16th Portuguese Conference on Artificial Intelligence, EPIA 2013, Sep 2013, Azores, Portugal. pp.114-125, ⟨10.1007/978-3-642-40669-0_11⟩ |
|
|
Automatic segmentation of hippocampus in histological images of mouse brains using deformable models and random forest25th IEEE International Symposium on Computer-Based Medical Systems (CBMS’12), Jun 2012, Rome, Italy. pp.1-4, ⟨10.1109/CBMS.2012.6266318⟩ |
|
|
A Comparative Study of Three GPU-Based Metaheuristics12th International Conference on Parallel Problem Solving from Nature (PPSN’12), Sep 2012, Taormina, Italy. pp.398-407, ⟨10.1007/978-3-642-32964-7_40⟩ |
|
|
Bayesian neural network priors at the level of unitsBayesian Statistics in the Big Data Era, Nov 2018, Marseille, France. pp.1 |
|
|
Texture Classification of Proteins Using Support Vector Machines and Bio-inspired MetaheuristicsBiomedical Engineering Systems and Technologies, 452, pp.117-130, 2014, 978-3-662-44485-6. ⟨10.1007/978-3-662-44485-6_9⟩ |
|
|
Bayesian neural networks increasingly sparsify their units with depth2018 |
|
|
AUTOMATIC SEGMENTATION OF ANATOMICAL STRUCTURES USING DEFORMABLE MODELS AND BIO-INSPIRED/SOFT COMPUTINGArtificial Intelligence [cs.AI]. University of Parma, 2014. English. ⟨NNT : ⟩ |