TENNs-PLEIADES: Building Temporal Kernels with Orthogonal Polynomials
Introduces PLEIADES (PoLynomial Expansion in Adaptive Distributed Event-based Systems), a parameterization using orthogonal polynomials that enables Temporal Event-Based Neural Networks to classify event-sensor data directly in the spatio-temporal domain. Demonstrates state-of-the-art results in hand gesture recognition, eye tracking, and PROPHESEE’s 1-Megapixel Automotive Detection dataset, all running on Akida 2.0 IP.
Want More Research Like This?
Full access to tools, docs, models, and all future white papers. Free to join. No spam. Unsubscribe anytime.
