Lionel Lacassagne -

Registered user since Mon 29 Oct 2018

Name:Lionel Lacassagne

Bio:

Lionel LACASSAGNE is Professor in Computer Science at Sorbonne University (Formerly University Pierre et Marie Curie - Paris 6). He got an Engineering Degree in Computer Science in 1995 (EPITA) and a Doctorate in Robotics in 2000 (Paris 6 University). He has been assistant professor (2002–2015) in Paris-Sud University. He is an expert in code optimization for SIMD multi-core processors, GPUs, Cell processor and VLIW DSP. His current research areas deal with high-performance computing and image processing applied to embedded systems. His current researches focus on the design of new algorithms for multi-core, many-core processors and GPUs.

Country:France

Affiliation:Sorbonne University — LIP6

Personal website:http://www-soc.lip6.fr/~lacas/

Research interests:Computer Architecture, SIMD, GPU, Computer Vision, High-Performance Embedded System (eHPC), System on Chip, design of new irregular algorithm tuned for parallel architecture, Connected Component Labeling, Connected Component Analysis

Contributions

WPMVP 2020	Author of Welcome Talk within the WPMVP 2020-track Session Chair of Main session (part of WPMVP 2020) Committee Member in Organizing Committee within the WPMVP 2020-track Author of How to speed Connected Component Labeling up with SIMD RLE algorithms within the WPMVP 2020-track Author of Open Discussion within the WPMVP 2020-track
WPMVP 2019	Author of Open Discussion within the WPMVP 2019-track Session Chair of WPMVP Session 2 (part of WPMVP 2019) Session Chair of WPMVP Session 1 (part of WPMVP 2019) Organizing Chair in Organizing Committee within the WPMVP 2019-track Author of Designing efficient SIMD algorithms for direct Connected Component Labeling within the WPMVP 2019-track Author of Welcome to WPMVP workshop within the WPMVP 2019-track
WPMVP 2018	Committee Member in Program Committee within the WPMVP 2018-track Author of Usuba, Optimizing & Trustworthy Bitslicing Compiler within the WPMVP 2018-track Author of Small SIMD Matrices for CERN High Throughput Computing within the WPMVP 2018-track
WPMVP 2016	Presenter of A new SIMD iterative connected component labeling algorithm within the WPMVP 2016-track Committee Member in Program Committee within the WPMVP 2016-track