Vision System on Module (SoM)

tinyVision.ai has addressed the challenge of integrating CV into IoT devices using a tightly integrated platform consisting of a low power image sensor, IMU and microphone coupled to a local computing device (Lattice ultra low power ICE40 5K FPGA). The modules are designed to be dropped into end products as-is with no Hardware modifications, significantly shortening the time to market of the end product.

The SoM enables IoT devices to see, hear and sense motion.

The module integrates the following hardware capabilities:

Processing tile	Lattice ICE40 Ultra-low-power FPGA, 5K LUT, 1Mb RAM, 8 MAC units
Image sensor options	Himax HMB010 (Color, rolling shutter)
Audio	One I2S microphone, expandable with off-board microphone
LED's	Tri-colour LED IR LED for low light illumination
Motion	6 axis Invensense Gyro/accelerometer
GPIO	Four, programmable IO voltage
Memory	8Mb qSPI Flash for bitstream/code storage 64Mb qSPI SRAM for data
Host Interface	4 wire SPI, programmable voltage levels
Power	2 power options: Single 3.3V operation, can supply 1.8V and 1.2V @100mA (max) to external devices using onboard LDO External 3.3V, 1.8V, 1.2V for lower power operation
SW support	Lattice SensAI (Tensorflow/Caffe/Keras supported for model development, toolchain for quantization and mapping to Neural Network engines.

 

This product is in development and will be available through