About me. I'm a Computer Vision Research Scientist (PhD) currently working at SLAMcore Limited. I'm specialized in long term, large scale and efficient SLAM technologies. I'm also partly working on dense mapping and deep learning to improve the spatial AI stacks like Pose Tracking, Mapping and Semantics
I did my PhD at the Université Clermont Auvergne, in Clermont-Ferrand, France in partnership with the ENCOV research group at the Institut Pascal and the LVIC laboratory at CEA-LIST under the supervision of Professor Adrien Bartoli and Mohamed Tamaazousti . I received my master's Degree in computer science from the EPITA in Paris, France, in 2013. The research objectives of the thesis were to model the light sources to improve the rendering of augmented reality applications.
Current situation. Open for full remote (international) or partially onsite (Paris) opportunities. You can contact me on my Linkedin or directly at morgand.alexandre_at_ gmail.com. You can find my resume here.

Publications

Un modèle géométrique multi-vues des taches spéculaires basé sur les quadriques avec application en réalité augmentée
Alexandre Morgand
Thesis report (French)
A Multiple-View Geometric Model of Specularities on Non-Uniformly Curved Surfaces
Alexandre Morgand, Mohamed Tamaazousti, Adrien Bartoli
VRST 2017 (Poster)
Poster of John VRST 2017
Color Consistency of Specular Highlights in Consumer Cameras
To increase the realism in rendering, estimating the illumination in the scene is a key element. A lot of works tackle this problem but rarely discuss the color of the reconstructed illumination. The Dichromatic Model indicates that the specular component is not affected in color by the texture underneath and holds the light source’s color. Though theoretically sound, in practice consumer cameras are subject to nonlinear behaviors which change RGB ra- tios and create inconsistencies when estimating the illumination. In this paper, we study the conditioning and limits of inverting local illumination models while relying on the Dichromatic Model. We show that the reconstructed specular component has an inconsis- tent color because it changes depending on the surface’s colors.
John Lin, Mohamed Tamaazousti, Souheil Hadj Said, Alexandre Morgand
VRST 2017 (Poster)
A Multiple-View Geometric Model of Specularities on Non-Planar Shapes with Application to Dynamic Retexturing
We propose dual JOLIMAS, a new model which raises the planarity assumption. It uses the fact that specularities remain elliptical on convex surfaces and that every surface can be divided in convex parts. The geometry of dual JOLIMAS then uses a 3D quadric per convex surface part and light source, and predicts the specularities by a means of virtual cameras, allowing it to cope with surface’s unflatness.
Alexandre Morgand, Mohamed Tamaazousti, Adrien Bartoli
TVCG 2017 special issue: selected papers from ISMAR 2017 (Journal/Oral) - Acceptance rate 12.4%
HTML5 Icon
Modèle géométrique multi-vues de spécularité
Alexandre Morgand, Mohamed Tamaazousti, Adrien Bartoli
Orasis 2017 (Oral) - June 2017
A Geometric Model for Specularity Prediction on Planar Surfaces with Multiple Light Sources
This paper highlights our empirical model called JOLIMAS (JOint LIght-MAterial Specularity) which is used to predict specularities in real-time on planar surfaces. We used this model to perform retexturing on several objects. The novelty of paper compared to our previous approach is the handling of multiple light sources of different intensities. Moreover, light sources can be turned on and off.
Alexandre Morgand, Mohamed Tamaazousti, Adrien Bartoli
TVCG 2017 (Journal) - Acceptance rate 20%
An Empirical Model for Specularity Prediction with Application to Dynamic Retexturing
This paper highlights our empirical model called JOLIMAS (JOint LIght-MAterial Specularity) which is used to predict specularities in real-time on planar surfaces. We used this model to perform retexturing on several objects.
Alexandre Morgand, Mohamed Tamaazousti, Adrien Bartoli
ISMAR 2016 (Oral) - Acceptance rate 22.5%
Modèle empirique de prédiction géométrique de spécularité : application au retexturing
Alexandre Morgand, Mohamed Tamaazousti, Adrien Bartoli
RFIA 2016 (Oral) - acceptation rate 25%
HTML5 Icon
Reconstruction d'une source lumineuse modélisée par une quadrique à partir d'images multiples
Alexandre Morgand, Mohamed Tamaazousti, Adrien Bartoli
Orasis 2015 (Oral)
HTML5 Icon
Détection générique et temps réel des spécularités
Alexandre Morgand, Mohamed Tamaazousti, Adrien Bartoli
Orasis 2015 (Oral)
HTML5 Icon
Generic and real-time detection of specular reflections in images
We proposed a real-time method for specularity prediction. The novelty of the approach was the use of a pre-treatment step to level the brightness of the image, a thresholding process according to the brightness of the image and a post-processing step which use the gradient aspect of a specularity.
Alexandre Morgand, Mohamed Tamaazousti
VISAPP 2014 (Oral)

Patents

HTML5 Icon
Image Processing in the Presence of Specularities (Procédé de traitement d'images avec spécularités)
Mohamed Tamaazousti, Alexandre Morgand, Adrien Bartoli
Patent registered in june 2016