Motion and Tracking Capability

NeuronBasic’s SoC is capable of gesture motion identification. The target gesture poses could be programmed according to actual application requirements, such as specific gesture poses or arm waving direction. Each target movement input can trigger a specific signal to enable pre-programmed functions. These functions could include power on/off switching, status LED flashing, audio playback, volume increase/decrease, or previous/next function buttons….etc. All the functions could be activated without the need to touch any button or needing to connect to the internet for background computing. They could be widely applied in traditional speakers, smart speakers or other existing accessories that could benefit from an AI upgrade.

In addition, the SoC could also used for tracking purpose. As long as the product design supports motor running features on the device to turn or pan, the algorithm can perform tracking of the desired target with the required movements. If the target is set to be a pedestrian, then wherever the person goes the device will pan, following the trace made by the pedestrian when he or she moved. It could be applied as a single feature or combined with the other features that the original design could support, such as image capture, toy drone following and the other tracking functions.

Bluetooth Speaker Application

• Timer: Reminder for event / Report Time
• Touchless Control: Previous / Next tMP3 track, volume control by gesture
• Voice Memo Recorder: by Face or Pedestrian detection
• Watch Dog: Welcome, Caution or warning trigger by Pedestrian detection

• Better false detection performance with moving backgrounds
• Leverage existing battery, memory, mic and speaker in existing designs
• Add-on value for normal BT speakers - achieve better product pricing