Jehan-Antoine Vayssade (30 years old)
Postdoc in animal tracking trought computer vision
PhD in computer science, specialized in image analysis
Master thesis about the monitoring of small ruminants
High skills in computer graphics (real time rendering)
- This email address is being protected from spambots. You need JavaScript enabled to view it.
About
During my adolescence, I developed a strong interest in video game development which led me to acquire a solid knowledge in various scientific domains related to computer science, including real-time rendering, optimization, 3D rendering engines, sockets, and operating systems. As I worked on building my own game 15 years ago, I discovered a deeper fascination with the underlying mechanics of game development, which drove me to develop my own complete game engine. This engine features a custom 3D engine, integrated physics, post-processing (shader), sound, scripting, editor, and server, all programmed in c++ 11. Through this self-learning journey, I have honed my problem-solving skills and increased my programming proficiency, making me more adaptable to new challenges. Eventually, I pursued a master's degree in Image Analysis and Computer Graphics.
After completing my master's degree in Image Analysis and Computer Graphics, I continued my academic journey and obtained a PhD in the same field. This provided me with a specialized expertise in deep learning and hyper-parameter optimization, applied to image analysis, including hyper-spectral data, image segmentation, object detection, and tracking. Through my research, I have developed a comprehensive understanding of advanced image analysis techniques and their practical applications. This knowledge has proven invaluable in my work as a computer scientist, allowing me to stay on the cutting edge of technological advancements and to provide innovative solutions to complex problems.
Feel free to follow me for more updates !
 |
 |
 |
 |
 |
 |
Ecology, biodiversity, and agriculture have been long-standing passions of mine, especially when it comes to the idea of doing things oneself and achieving a sense of autonomy. These interests were instilled in me from a young age and have shaped my personal projects, skills, and knowledge in this field. As a researcher, I have less time to dedicate to these pursuits, but I am always eager to engage in discussions surrounding these topics.
At first glance, the intersection of technology and autonomy may appear contradictory. However, I firmly believe in striking a balance between the two, keeping our head in the clouds while remaining grounded. While technology can be fascinating and enable incredible achievements, it is important to recognize its costs and the barriers to access it that exist for those who are economically disadvantaged. Thus, I believe that the best technology is that which is open source, free, and easily reproducible with a low energy cost and minimal ecological impact, in order to respect both nature and future generations.
Here you can see my works and studies, follow links to see some of my codes or associated ressources.
... good reading ...
Skills
- latex / tikz
- c / c++
- python 3
- php / sql / js
- linux / slackware
- irrlicht
- opengl / glsl
- computer vision
- tensorflow / torch
- image analysis
- 3d engine
- signal processing
- game-engine
- socket
- c# / unity
- java
- opencl / cuda
- github
- gitlab
- ...
Studies
| Dijon | Doctorat IT department (image analysis and deep-learning) | manuscript | |
| Toulouse | Master 1&2 IGAI IT department | 4nd manuscript | |
| Chambéry | Licence 3 IT department | 2nd | |
| Rodez | DUT IT department | ||
| Rodez | Bac Pro *SEN | mention "Bien" | |
| Rodez | Bep *SEN | 2nd |
*SEN : Electronic and Numeric Systems with Telecom and Network option
*IGAI : computer graphics and image analysis
Works
| 2022-2023 | Guadeloupe | research fellow / postdoc in computer vision at Inra Unité de Recherche en Zootechnie | |
| Dijon | doctorat in precision agriculture using computer vision on multi-spectral data at AgroSup/INRA/uB | ||
| Guadeloupe | internship python/opencl in computer vision at Inra Unité de Recherche en Zootechnie | ||
| Toulouse | master 2 project for rendering optimization at Irit (VORTEX/AGGA) | ||
| Toulouse | internship c++ in point cloud association at Irit (VORTEX/AGGA) | ||
| 2016-2017 | Montpeyroux | small professional website, magie de l'aubrac | |
| Rodez | developper c# unity at Spisoftware | ||
| Rodez | developper intranet php/mysql/ldap at ERDF | ||
| Decazeville | hardware / software enginer at Vanotec | ||
| 2008-2009 | Marcillac | production agent - domaine viticole de Cantagrel | |
| Pollestre | free hardware enginer at Emmaüs |
Research
| 28/08/2023 | EAAP | conference : Estimating sow posture from computer vision : influence of the sampling rate | co-writer |
| todo | INRAe | paper : Outdoor Goat Tracking in Natural Environments : A tracking dataset | todo |
| todo | INRAe | paper : Famacha estimation trought mobile-phone and deep-learning | writing |
| todo | INRAe | paper : Puzzle: unsupervised animal tracking | writing |
| 12/06/2023 | INRAe | presentation : Conseil Scientifique Inrae ASSET | speaker |
| 15/03/2023 | INRAe | paper : CherryChèvre: A Fine-Grained Dataset for Goat Detection in Natural Environments | published |
| 21/04/2023 | INRAe | presentation : Un petit regard sur l'astrophysique (simulation univers / modèle Janus) | speaker |
| 16/12/2022 | INRAe | paper : Wizard: unsupervised goats tracking algorithm | published |
| 08/03/2022 | Dijon | published | |
| 13/02/2022 | AgroSup | published | |
| 13/02/2022 | AgroSup | paper : Pixelwise Instance Segmentation Of Leaves In Dense Foliage | published |
| 01/07/2021 | Bordeau | poster | |
| 09/06/2021 | AgroSup | published | |
| 08/12/2020 | AgroSup | speaker | |
| 29/02/2020 | AgroSup | conference : International Conference on Computer Vision theory and Application (VISAPP) | talk |
| AgroSup | paper : Two-step multi-spectral registration via key-point detector and gradient similarity | published | |
| INRA | paper : Outdoor animal tracking combining neural network and time-lapse cameras | published | |
| INRA | short paper : automatic acitivity tracking using time-lapse cameras | published | |
| 26/06/2019 | INRA | conference : Goats monitoring at the pasture scale combining neural network and time-lapse cameras | published |
| 30/04/2019 | INRA | poster : Approche multicritère pour la caractérisation des adventices par imagerie | poster |
| INRA | paper : automatic activity tracking using drone | published | |
| home | game engine optimisation | standby | |
| home | rendering optimisation | standby | |
| home | multi-threading synchronisation without hardware lock | standby |
Self-learning and online followed courses
| 2022-2023 | Guadeloupe |
Les Journées Ouvertes en Biologie, Informatique et Mathématiques |
| Dijon-Lyon Belfort-Paris |
Formations doctorales diverses (enseignement, éthique, déontologie, vulgarisation, ...) |
|
| Toulouse |
Fluid Simulation |
|
| Toulouse | ||
| Chambéry |
soil characterization |
|
| at home |
iron melting / welding / tanning / soap |
Interest & project
- game like (m)morpg & game engine & server (closed src in dev)
- irrlicht addon
- Nenkan a tool to manage animes/mangas description and episodes list
- ShadowInstaller a tool (with gui or in silent mode) to install apps or do scripts
- 3D c++11 flexible engine
- montpeyroux website sketch
- moderator on irrlicht-fr.org (Magun)
- few interesting link