
Pablo Mesejo Santiago
Presentation
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
Conference papers
hal-04873693
v1
|
|
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⟩
Conference papers
hal-02882229
v4
|
|
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⟩
Conference papers
hal-02054236
v1
|
|
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
Conference papers
hal-02177151
v1
|
|
Bayesian neural networks become heavier-tailed with depthNeurIPS 2018 - Thirty-second Conference on Neural Information Processing Systems, Dec 2018, Montréal, Canada. pp.1-7
Conference papers
hal-01950658
v1
|
|
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⟩
Conference papers
hal-01851738
v1
|
|
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⟩
Conference papers
hal-01851511
v1
|
|
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
Conference papers
hal-01950659
v1
|
|
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⟩
Conference papers
hal-01504847
v1
|
|
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⟩
Conference papers
hal-01221126
v1
|
|
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⟩
Conference papers
hal-01221343
v1
|
|
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⟩
Conference papers
hal-01221328
v1
|
|
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⟩
Conference papers
hal-01221602
v1
|
|
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⟩
Conference papers
hal-01221570
v1
|
|
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
Conference papers
hal-01221613
v1
|
|
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⟩
Conference papers
hal-01221512
v1
|
|
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⟩
Conference papers
hal-01221660
v1
|
|
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⟩
Conference papers
hal-01221645
v1
|
|
Bayesian neural networks increasingly sparsify their units with depth2018
Preprints, Working Papers, ...
hal-01950657
v1
|
|
Bayesian neural network priors at the level of unitsBayesian Statistics in the Big Data Era, Nov 2018, Marseille, France. pp.1
Conference poster
hal-01950660
v1
|
|
AUTOMATIC SEGMENTATION OF ANATOMICAL STRUCTURES USING DEFORMABLE MODELS AND BIO-INSPIRED/SOFT COMPUTINGArtificial Intelligence [cs.AI]. University of Parma, 2014. English. ⟨NNT : ⟩
Theses
tel-01363683
v1
|
|
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⟩
Book sections
hal-01221496
v1
|