JeVois
1.22
JeVois Smart Embedded Machine Vision Toolkit
|
|
Updated: Setp 9, 2021
Feature | Status | Notes |
---|---|---|
Hardware Design | - | - |
All 19 internal power supplies operational from single 12-24 VDC input | - | |
4x A73 big cores running at full speed with no errors under full load | - | |
2x A53 little cores running at full speed with no errors under full load | - | |
4GB LPDDR4 working with no error at full DDR4-3200 speed | - | |
ARM NEON accelerated multimedia processor instructions tested and working | - | |
A311D MALI GPU OpenGL-ES 3.2 working using ARM demos (framebuffer mode) | - | |
Micro-SD / TF card working at SDR104 (104 MBytes/s) speed | - | |
HDMI output, video + sound | - | |
2x USB 2.0 type-A ports at top of camera provide power and are working | - | |
Mini-USB 2.0 OTG port working, both in host and in device modes | - | |
4-pin micro-serial port working | - | |
6-pin AUX power port working (outputs 5V, 3.3V, 1.8V for accessories) | - | |
8-pin GPIO port working (configurable as SPI+I2C, or general GPIO) | (12) | |
Thermally stable operation under full load | - | |
Camera sensor connector working, camera functionality | - | |
Camera sensor connector working, dual-camera functionality | ? | (10) |
Camera sensor connector working, SPI inertial measurement unit (IMU) functionality | - | |
M.2 E-Key connector working with PCIe M.2 cards (Coral TPU, Wifi cards) | - | |
M.2 E-Key connector working with USB/SDIO M.2 cards (Wifi/bluetooth cards) | - | |
Wired ethernet using add-on USB-to-RJ45 converter | - | |
Variable-speed cooling fan control | - | |
Base System | - | - |
Bootloader, Linux Kernel 4.9.x, partitions, Ubuntu 20.04 | - | |
Custom kernel device tree | - | |
Boot to X-windows and use as a regular computer when connecting keyboard/mouse | - | |
Play Youtube videos with sound over HDMI | - | |
Boot to console | - | |
Boot to JeVois software | - | |
JeVois-Pro video + serial + storage over mini-USB OTG, kernel Gadget driver | (13) | |
JeVois-Pro control via JeVois Inventor | (14) | |
Software reset | - | |
Camera Sensor | - | - |
Sony IMX-290 2MP camera sensor detection and basic 1080p @ 30fps operation | - | |
Sony IMX-290 use 4x MIPI-CSI lanes for lowest latency | - | |
Sony IMX-290 hardware-accelerated conversion from Bayer to YUYV, RGB, RGBA, GREY | - | |
Sony IMX-290 support for basic controls (brightness, exposure, gain, etc) | - | |
Sony IMX-290 frame cropping support, any size from native 1080p | (11) | |
Sony IMX-290 ISP-based hardware-accelerated frame scaling support, any size | (11) | |
Sony IMX-290 ISP-based dual-stream capture, e.g., YUYV 1920x1080 + RGB 512x288 | - | |
Sony IMX-290 native 720p support | (1) | |
Sony IMX-290 frame rate control | (2) | |
IMU Sensor | - | - |
TDK ICM-29048 inertial measurement uint (IMU) detection over 7MHz SPI bus | - | |
TDK ICM-29048 IMU basic operation, accel, gyro, compass, temperature over 7MHz SPI | - | |
TDK ICM-29048 IMU upload digital motion processor (DMP) code | - | |
TDK ICM-29048 IMU run digital motion processor (DMP) code | (3) | |
Machine Vision Base Software | - | - |
OpenCV 4.5.3 + OpenVino 2021.4 + all OpenCV-Contrib + Python3 bindings | - | |
OpenGL-ES 3.2 framebuffer | - | |
OpenCL 1.2 | - | |
Python 3.8 + numpy + scipy + any aarch64 python packages | - | |
TensorFlow Lite 2.5 | - | |
Mediapipe 0.8 python | - | |
Vulkan, hardware H.265 encode/decode, etc | ? | (4) |
JeVois Base Software | - | - |
JeVois use OpenGL for 1080p display @ 30fps over HDMI | - | |
JeVois OpenGL support for zero-copy video display using DMABUF, EGLImageKHR | - | |
JeVois OpenGL DMABUF, EGLImageKHR support for RGB, RGBA, YUYV, GREY direct display | - | |
JeVois OpenGL support for DMABUF display of YUYV, RGB, RGBA, GREY from camera | - | |
JeVois OpenGL support for display of processed YUYV, RGB, RGBA, GREY using shaders | (7) | |
JeVois capture fullsize+scaled streams from camera, display fullsize, process scaled | - | |
JeVois new on-screen graphical user interface using Dear ImGui 1.83 | - | |
JeVois new ImGui backend for MALI OpenGL-ES GPU in framebuffer mode | - | |
JeVois new drivers to capture console keyboard/mouse events and pass to ImGui | - | |
JeVois run OpenCV machine vision algorithms written in C++ | - | |
JeVois run other machine vision algorithms written in C++ | - | |
JeVois run OpenCV machine vision algorithms written in Python | - | |
JeVois run other machine vision algorithms written in Python | - | |
JeVois support for Python modules to draw OpenGL overlays (box, text, etc) | - | |
JeVois-Pro run machine vision algorithms developed for JeVois-A33, C++ and Python | - | |
JeVois-Pro enhance some JeVois-A33 algorithms to use DMABUF zero-copy + OpenGL | - | |
JeVois new dual threadpool with user-selectable big/little affinity | - | |
JeVois new async_big() and async_little() thread launching functions | - | |
JeVois New DNN Software | - | - |
New JeVois deep neural network (DNN) framework | - | |
JeVois DNN framework support for classification, detection, semantic segmentation | - | |
JeVois DNN framework hierarchical model zoo with 30+ pre-trained models | - | |
JeVois DNN framework download and run custom models | - | |
JeVois DNN run models on CPU using OpenCV DNN module | - | |
JeVois DNN run models on A311D 5-TOPS integrated neural processing uint (NPU) | - | |
JeVois custom kernel driver to enable Coral 4-TOPS accelerator over PCIe-2.1 (5 Gbps) | - | |
JeVois DNN run models on Coral 4-TOPS accelerator, PCIe-2.1 (5 Gbps) M.2 card | - | |
JeVois DNN run models on Coral 4-TOPS accelerator, USB-2.0 (480 Mbps) dongle | - | |
JeVois DNN run models on Myriad-X 1-TOPS accelerator with OpenVino, PCIe-2.1 M.2 card | - | |
JeVois DNN run models on Myriad-X 1-TOPS accelerator with OpenVino, USB-2.0 dongle | - | |
JeVois DNN run models on CPU, A311D NPU, Coral TPU, Myriad-X VPU in parallel | - | |
JeVois DNN run multiple models on one CPU, NPU, TPU, VPU hardware, time multiplexed | - | |
Support for Coral M.2 board with 2xTPU chips using 2xPCIe lanes, made by Google | (5) | |
Support for custom JeVois 2xTPU board with 1 TPU on PCIe and the other on USB | - | |
Support for custom JeVois 2xTPU + eMMC board with PCIe hub + 2 TPUs on PCIe | - |