Recently, uSens officially launched a globally leading gesture recognition system solution based on the Rockchip Toybrick TB-RK3399ProD development platform, which can be widely applied in fields such as artificial intelligence, AIoT, intelligent vision, smart voice, new retail, and personal consumer electronics, attracting much attention. This follows uSens’s technical upgrade after the debut of the gesture recognition solution based on the Rockchip RK1808 chip at the Rockchip Developer Spring Conference in April, meeting the diverse demands of the market for product performance and applications.
The Toybrick TB-RK3399ProD development platform adopts the big.LITTLE CPU architecture, featuring a dual-core Cortex-A72 and a quad-core Cortex-A53, demonstrating technical leadership in overall performance and power consumption; the quad-core ARM high-end GPU Mali-T860 integrates more bandwidth compression technologies, offering excellent overall performance, and its on-chip NPU (Neural Processing Unit) achieves a computational performance of up to 3.0 Tops. It has rich external interfaces, supporting PCIE×4 + Mini PCIE, dual USB 3.0 high-speed interfaces Type-C + USB 3.0 Type-A, dual MIPI CSI, and dual ISP, with pixel processing capabilities of up to 1300W pixels supporting HDMI 2.1, DP 1.2 MIPI-DSI, EDP, and other display outputs; it also supports an 8-channel digital microphone array input. On the software side, it comes pre-installed with Android and Linux systems, supporting dual system boot and one-click switching; it integrates a rich set of AI application development components, providing a friendly and simple application programming interface, seamlessly connecting with the NPU to achieve AI computation hardware acceleration, supporting various model inferences such as TensorFlow/TensorFlow Lite/Caffe.
uSens has launched two gesture recognition solutions based on the Toybrick TB-RK3399ProD development platform, namely the RGB Gesture Recognition Control System and the Infrared Binocular FINGO Hardware 3D Gesture Solution. Among them, the RGB Gesture Recognition Control System’s core technology identifies 22 key points on the hand and 26 degrees of freedom to discern hand poses, achieving full gesture recognition. The system is mature and stable, featuring the following highlights:
1. Low environmental requirements: accurate gesture recognition under normal lighting conditions
2. Wide recognition distance range: the distance between the hand and the camera can be 10cm-80cm
3. Low latency: within 15ms
4. Easy installation method: download the uSens gesture APK application package, install it directly to use
5. Simple interface: real-time display of background images and 3D skeletal information of gestures
6. High cost-performance ratio
Demonstration video of the RGB gesture recognition control system based on RK3399Pro
The Infrared Binocular FINGO hardware 3D gesture solution combines uSens’s infrared binocular hardware FINGO with RK3399Pro, equipped with two infrared cameras, allowing gestures to have depth information and enabling tracking of 3D gestures, greatly expanding the application range, meeting the needs of users who require 3D gestures, applicable in smart homes, automotive, security, and other fields.
Infrared binocular FINGO hardware based on RK3399Pro
Demonstration video of the 3D gesture solution
Earlier, uSens launched the RGB gesture recognition solution based on the Rockchip RK1808 neural network processor, primarily targeting the home appliance sector. By applying the uSens gesture recognition solution based on RK1808 on smart air conditioners and televisions, gesture control can replace physical remote controls, greatly enhancing the convenience of smart home operations.
Demonstration video of the RGB gesture recognition solution based on RK1808
The Rockchip Toybrick development platform integrates software and hardware development, greatly assisting solution providers in accelerating the R&D process of industry applications in the field of artificial intelligence. Currently, in addition to the TB-RK3399ProD development platform, it has also launched the RK1808 AI computing stick with powerful deep learning inference capabilities. Through the ultra-high computing power of the computing stick, embedded devices can build artificial intelligence algorithms at the network edge, enabling traditional embedded devices to easily upgrade to artificial intelligence.Products on the Toybrick platform can be purchased at the official corporate store:https://dwz.cn/SrMK19s1.
uSens has leading gesture control technology globally, and the gesture recognition solutions based on the Rockchip Toybrick RK3399Pro and RK1808 platforms will greatly expand the application of gesture control in more scenarios, promoting the transformation of human-computer interaction methods.
How to Find Me
1. Click the blue text at the top Shenzhen Automotive Electronics Industry Association to follow;
2. Search for the WeChat public account: “Shenzhen Automotive Electronics Industry Association” or “qidianxiehui”;
3. Scan the QR code to follow
