Our researchers

Antoine Manzanera

Unité d'Informatique et d'Ingénierie des Systèmes (U2IS)

Professeur

HDR

I am Professor of Computer Science at ENSTA Paris and Researcher in Computer Vision within the Autononous Systems and Robotics Laboratory.

I graduated from Université Claude Bernard in Lyon (BSc Mathematics, 1991 and MSc Theoretical Computer Science, 1993), Télécom ParisTech (PhD Signal and Image Processing, 2000) and Université Pierre et Marie Curie (Habilitation Degree, 2012).

Research and education activities

For 20 years now I have taught Image Processing, Discrete Geometry, Mathematical Morphology, Computer Vision and Machine Learning in ENSTA, Université Paris Sorbonne, Université Paris Saclay, and in different Universities around the world.

My Research interests extend across the broad range of mathematical, physical and technological foundations of Computer Vision, with a particular focus on Video Analysis, Image Representations, Scene Reconstruction, and Autonomous Visual Learning.

Our researchers

Antoine Manzanera

Research and education activities

HAL Publications

A Mixed audio-video SPD network for online classification of Parkinsonian speech patterns

A Riemannian multimodal representation to classify parkinsonism-related patterns from noninvasive observations of gait and eye movements

A Novel Hybrid Grid Search and Tree Parzen Estimator for Deep Learning Hyperparameters Optimization

Leveraging a realistic synthetic database to learn Shape-from-Shading for estimating the colon depth in colonoscopy images

Estimating Polyp Size From a Single Colonoscopy Image Using a Shape-From-Shading Model

NECO: NEURAL COLLAPSE BASED OUT-OF-DISTRIBUTION DETECTION

InfraParis: A multi-modal and multi-task autonomous driving dataset

Improving Knee Osteoarthritis Classification with Markerless Pose Estimation and STGCN Model

Riemannian SPD learning to represent and characterize fixational oculomotor Parkinsonian abnormalities

Does it work outside this benchmark? Introducing the rigid depth constructor tool

Rebalancing gradient to improve self-supervised co-training of depth, odometry and optical flow predictions

Hyperparameter Optimization of Deep Learning Models for EEG-Based Vigilance Detection

An Oculomotor Digital Parkinson Biomarker From a Deep Riemannian Representation

A Study of Deep Perceptual Metrics for Image Quality Assessment

A multimodal Parkinson quantification by fusing eye and gait motion patterns, using covariance descriptors, from non-invasive computer vision

Quantification of Parkinsonian Kinematic Patterns in Body-Segment Regions During Locomotion

Learning Graph Representation with Randomized Neural Network for Dynamic Texture Classification

A recurrent approach for predicting Parkinson stage from multimodal videos

Does it work outside this benchmark? Introducing the Rigid Depth Constructor tool, depth validation dataset construction in rigid scenes for the masses.

Naturally Constrained Online Expectation Maximization

A compact and recursive Riemannian motion descriptor for untrimmed activity recognition

UVP6: a new low power, low cost and deep ocean rated imaging sensor for automatic quantification of particles and plankton from autonomous platforms

A Fractal-Based Approach to Network Characterization Applied to Texture Analysis

Integrating Experimental Data Sets and Simulation Codes for Students into a MOOC on Aerial Robotics

DroMOOC

Bio-inspired Perception Sensor (BIPS) Concept Analysis for Embedded Applications

Learning structure-from-motion from motion

DroMOOC: a Massive Open Online Course on Drones and Aerial Multi Robot Systems

Spatio-temporal multi-scale motion descriptor from a spatially-constrained decomposition for online action recognition

Dual field combination for unmanned video surveillance

End-to-end depth from motion with stabilized monocular videos

Fast Semi Dense Epipolar Flow Estimation

Mixing Hough and Color Histogram Models for Accurate Real-Time Object Tracking

Multi range Real-time depth inference from a monocular stabilized footage using a Fully Convolutional Neural Network

The Visual Object Tracking VOT2017 challenge results

A coronary artery segmentation method based on multiscale analysis and region growing

Action-centric Polar Representation of Motion Trajectories for Online Action Recognition

Automatic Bifurcation Detection in Coronary X-Ray Angiographies

Fast growing hough forest as a stable model for object detection

Line and circle detection using dense one-to-one Hough transforms on greyscale images

Statistical binary patterns for rotational invariant texture classification

Topological Attribute Patterns for Texture Recognition

A Gaussian mixture representation of gesture kinematics for on-line Sign Language video annotation

Feature Extraction on Local Jet Space for Texture Classification

A versatile object tracking algorithm combining Particle Filter and Generalised Hough Transform

Noise Tolerant Descriptor for Texture Classification

Transformées de Hough denses : un cadre efficace et générique pour la reconnaissance de formes paramétrées

Automatic analysis and characterization of the hummingbird wings motion using dense optical flow features

Incorporating two first order moments into LBP-based operator for texture categorization

Video++, a Modern Image and Video Processing C++ Framework

Impact of topology-related attributes from Local Binary Patterns on texture classification

Representing activities with layers of velocity statistics for multiple human action recognition in surveillance applications

Spatial motion patterns: action models from semi-dense trajectories

Video Extruder: a semi-dense point tracker for extracting beam of trajectories in real time

Action recognition using bag of features extracted from a beam of trajectories

Motion Trend Patterns for Action Modelling and Recognition

Revisiting LBP-Based Texture Models for Human Action Recognition

Dense Hough transforms on gray level images using multi-scale derivatives

En mouvement, du modèle à la puce : Pour des systèmes de vision polyvalents et performants.

Analysing the hovering flight of the hummingbird using statistics of the optical flow field

A Motion Descriptor Based on Statistics of Optical Flow Orientations for Action Classification in Video-Surveillance

Detection of floating mines in infrared sequences by multiscale geometric filtering

Exact and Approximate Median Splitting on Distributed Memory Machines

Real Time Semi-dense Point Tracking

Local Jet Feature Space Framework for Image Processing and Representation

OBJECT MODELLING, DETECTION AND LOCALISATION IN MOBILE VIDEO: A STATE-OF-THE-ART

Characterization of motion cardiac patterns in magnetic resonance cine

Classification of Traversable Surfaces for Navigation Combining Multiple Cues Toward a Visual-Based Learning System

Extracción y Seguimiento de los Miembros Inferiores sin Marcadores

Image representation and processing through multiscale local jet features

Local Jet Based Similarity for NL-Means Filtering

Automatic Analysis of Microscopic Images in Hematological Cytology Applications

Image Characterization from Statistical Reduction of Local Patterns

High performance motion detection: some trends toward new embedded architectures for vision systems

Evolution of visual controllers for obstacle avoidance in mobile robotics

Motion detection: Fast and robust algorithms for embedded systems