Better Beat Tracking
The beat tracker your LED props deserve, a newly labeled 90 hour dataset, and TinyML that’s actually tiny.
This website is all human made. RSS/Atom
Big Posts
Fun with Local Image Features
Multi Kernel Descriptors, a different take on a DoG blob detector, and trying to make it all fast using Vulkan compute.
Bite Sized
Using the Data Cache on Cortex M7 STM32H7 (Rust)
Without breaking DMA peripherals that is
Machine Learning Inference on STM32/Cortex M7
tl;dr: Tensorflow is currently the way to go to run a QAT quantized model on Cortex Mx using optimized CMSIS-NN kernels.