VORA Ai 3399 Kit and Camera module(Not included).
VORA 3399 Kit.
This is ARM Board for Edge A.I. Development.
Since the NPU for A.I. calculation is built-in,
it is possible to develop A.I.products without additional hardware.
Rockchip 6-Core
RK3399
A.I. NPU
SPR2803S
Gigabit Ethernet
1Gbps
Wi-Fi
2.4/5GHz
Bluetooth
5.0 BLE
General Ubuntu
Linux
Supports General Ubuntu 20.04 LTS,
Not Embedded Linux.
It supports General Ubuntu Linux for the most A.I. development.
Additionally it supports various OSs such as Debian, Android, and Chromium OS.
of the Development. by the Development.
and, for the Development.
It supports GPIO, UART, I2C additional ports and
USB 2.0/3.0 ports for the development.
Also, it uses the same powering method as the computer (power button).
4K HDMI Out, MicroSD supported.
100dBA SNR, 24bit ALC5651 Audio Codec.
Use it in a variety of ways with the HDMI 2.0 smoothly
supports 4K resolution and additional functions.
The ability of RK3399, which combines ARM Cortex-A72
dual core and A53 quad core, also plays a role.
Lightspeeur 2803S AI NPU.
with Less than 1W of Power.
Lightspeeur SPR2803S NPU consumes
0.7W and performs 16.8TOPS high-calculation speed.
You can use VGG, MobileNet, ResNet model
and TensorFlow, PyTorch.
Infer Over 100 Images per Second¹.
We provides libraries and example code for development based on the built-in A.I. chip.
¹ Self-test criteria, Classification, 138 images per second.
5-Examples provided by XSLAB.
Classification², GNetSegmentation³,
GNetDetection⁴,
Face Recognition, and CDVA⁵.
There's nothing this little chip can't do.
² Single object recognition, ³ Multiple object recognition requiring precise boundaries,
⁴ Multi-object recognition without precise boundaries, ⁵ Compact Descriptor for Video Analysis
The photo is an example. It may differ from the actual running content or interface.
Small but
Powerful Experience.
Light and fast Xfce desktop GUI.
Also supports Korean and English.
Raspberry Pi Camera Module.
Another Power.
Sold separately. It can be used as is by selecting a terminal compatible with the camera for Raspberry Pi.
VORA Ai 3399 Kit
SPEC
VORA Ai 3399 Kit
Form Factor
Board (W145 x H145 x D65, mm)
Processor
Rockchip Inc. RK3399,
ARM® Cortex-A72 1.8GHz 64bit 2-core
ARM® Cortex-A53 1.4GHz 64bit 4-core
NPU
Gyrfalcon Inc. Lightspeeur® 2803S,
16.8TOPS @ 0.7W
Memory
LPDDR4 1866MHz 4GB On-board
GPU
ARM® Mali-T860 600MHz MP4
Storage
micro-SD
Image,
Pixel I/O
Image : 448*448, 224*224, 112*112,
56*56, 28*28, 14*14 pixel
Pixel/Layer : up to 4,014,080
16, 32, 64, 128, 256, 512, 1024 ch.
I/O Ports
2x USB 2.0, 2x USB 3.0,
20-pin(2.54mmm) Header Socket, GND,
GPIO, IC2(SCL, SDA), UART, micro-SD
Environment
Above Python 3.8
Above OpenCV 4.2
GNU
Network
1Gbps Ethernet, Wi-Fi 2.4/5GHz Bluetooth 5.0 BLE
Power Supply
12V/2A, DC
Frameworks
TensorFlow, PyTorch(Caffe),
VGG16, ResNet 18/50, MobileNet
Others
HDMI 2.0 (4K @60Hz), Realtek Inc. ALC5651
Raspberry Pi Camera Compatiable,
Temperature Sensor
OS
Ubuntu 20.04 + Xfce(GUI)
| SPEC | VORA Ai 3399 Kit |
|---|---|
| Form Factor | Board (W145 x H145 x D65, mm) |
| Processor | Rockchip Inc. RK3399, ARM® Cortex-A72 1.8GHz 64bit 2-core ARM® Cortex-A53 1.4GHz 64bit 4-core |
| NPU | Gyrfalcon Inc. Lightspeeur® 2803S, 16.8TOPS @ 0.7W |
| Memory | LPDDR4 1866MHz 4GB On-board |
| GPU | ARM® Mali-T860 600MHz MP4 |
| Storage | micro-SD |
| Image, Pixel I/O |
Image : 448*448, 224*224, 112*112, 56*56, 28*28, 14*14 pixel Pixel/Layer : up to 4,014,080 16, 32, 64, 128, 256, 512, 1024 ch. |
| I/O Ports | 2x USB 2.0, 2x USB 3.0, 20-pin(2.54mmm) Header Socket, GND, GPIO, IC2(SCL, SDA), UART, micro-SD |
| Environment | Above Python 3.8 Above OpenCV 4.2 GNU |
| Network | 1Gbps Ethernet, Wi-Fi 2.4/5GHz Bluetooth 5.0 BLE |
| Power Supply | 12V/2A, DC |
| Frameworks | TensorFlow, PyTorch(Caffe), VGG16, ResNet 18/50, MobileNet |
| Others | HDMI 2.0 (4K @60Hz), Realtek Inc. ALC5651 Raspberry Pi Camera Compatiable, Temperature Sensor |
| OS | Ubuntu 20.04 + Xfce(GUI) |