M. Tremblay
Papers
2
Total Citations
20
H-Index
2
About
M. Tremblay’s research lies at the intersection of neuromorphic vision, analog VLSI, and real-time machine perception. Their pioneering work on motion vision sensor architectures introduced a novel asynchronous self-signaling pixel design that detects moving edges directly on the focal plane, eliminating the need for frame-based processing. This approach, detailed in their 2002 paper (13 citations), enables ultra-low-latency motion computation by correlating temporal edge events in a dedicated coprocessing module—a foundational contribution to event-driven vision systems. Tremblay further advanced the field with a hexagonal CMOS sensor architecture (7 citations) that integrates analog edge extraction directly into the pixel array, supporting multiscale pattern recognition for robotic vision at resolutions up to 512×512 pixels. By embedding analog preprocessing on the sensor itself, Tremblay’s work dramatically reduces power consumption and data bandwidth, making real-time vision feasible for resource-constrained platforms. Their innovations in focal-plane processing and hexagonal addressing have influenced subsequent designs in smart cameras and autonomous navigation systems, cementing Tremblay’s role as a key architect of bio-inspired, energy-efficient visual sensing.
Research Focus
Key Achievements
Top Papers
- 1Motion vision sensor architecture with asynchronous self-signaling pixels13 citations · 2002
- 2