Thierry Bernard

École Nationale Supérieure de Techniques Avancées

Papers

Total Citations

H-Index

About

Thierry Bernard is a specialist in intelligent imaging systems and embedded vision processing, with a particular focus on programmable artificial retinas and focal plane array (FPA) architectures. His research sits at the intersection of CMOS sensor design, SIMD parallel processing, and autonomous robotics, exploring how computational intelligence can be embedded directly at the pixel level to enable smarter, more efficient visual processing. Bernard's most notable contribution lies in the development and advancement of Programmable Artificial Retinas (PARs) — CMOS imagers in which each pixel is equipped with a miniature digital processor, enabling on-sensor image analysis rather than raw data transmission. This paradigm shift reduces bandwidth demands and accelerates real-time visual tasks. His 2002 work on output methods for associative operations in these devices demonstrated how FPAs could deliver highly condensed, meaningful information rather than conventional image streams. Complementing this, his research on the VAMPIRE platform — a miniature retina-guided autonomous ground vehicle — illustrated the practical robotics applications of such embedded vision systems. His contributions to semistatic shift register design further refined the programmable retina architecture. While his citation counts remain modest, his work represents pioneering contributions to the field of neuromorphic and embedded vision computing.