Search papers, labs, and topics across Lattice.
MAGMA introduces a novel FPGA architecture that enables concurrent Gaussian Mixture Model (GMM) inference and online Expectation-Maximization (EM) parameter adaptation, addressing the limitations of fixed parameters in traditional GMM accelerators. By leveraging hardware-friendly transcendental approximations and a robust background update engine, MAGMA achieves significant performance improvements, including an 11.8脳 speedup in inference and an 81脳 speedup in the M-step over software implementations. The architecture demonstrates enhanced adaptability to non-stationary scenes, achieving a mean pixel accuracy of 81.5% compared to 79.7% for static models, showcasing the feasibility of online GMM learning on commodity edge FPGAs.
Achieving 11.8脳 faster inference and 81脳 faster adaptation on edge FPGAs, MAGMA revolutionizes GMM performance in dynamic environments.
Conventional FPGA-based Gaussian Mixture Model (GMM) accelerators use offline-trained, fixed parameters, limiting their ability to adapt to evolving scene statistics in long-lived edge systems. We present MAGMA, a fully synthesizable fixed-point FPGA architecture that performs concurrent GMM inference and online Expectation-Maximization (EM) parameter adaptation from a streaming RGB pixel input. MAGMA combines a pipelined inference datapath with a background update engine using hardware-friendly transcendental approximations---a range-reduced Chebyshev exponential, a CLZ-based logarithm, and a shift-and-subtract divider---alongside guards against variance collapse and cluster death that stabilize online fixed-point EM. Implemented on an AMD Spartan-7 XC7S50 with $K=4$ clusters, MAGMA runs at 74.49~MHz using 7,779 LUTs, 91 DSPs, and no block RAM, consuming 274~mW. It achieves an $11.8\times$ inference speedup and an $81\times$ M-step speedup over software, while spatial subsampling reduces per-update pixel volume by $40\times$ with minimal impact on EM convergence. Under a synthetic non-stationary scene, MAGMA's online adaptation improves mean pixel accuracy over a static baseline (81.5\% vs.\ 79.7\%), demonstrating that full online GMM learning is achievable on a commodity edge FPGA.
