Enabling energy efficient machine learning on a Ultra-Low-Power vision sensor for IoT
This work provides a domain-specific pipeline for computer vision applications in IoT, enabling the use of energy-efficient, low-power vision sensors despite their perception limitations.
This paper addresses the challenge of enabling energy-efficient machine learning on ultra-low-power vision sensors for IoT applications. The authors developed and implemented a real-time detection, classification, and tracking pipeline that achieves a power consumption of 7.5 mW for inference, which takes 8ms.
The Internet of Things (IoT) and smart city paradigm includes ubiquitous technology to extract context information in order to return useful services to users and citizens. An essential role in this scenario is often played by computer vision applications, requiring the acquisition of images from specific devices. The need for high-end cameras often penalizes this process since they are power-hungry and ask for high computational resources to be processed. Thus, the availability of novel low-power vision sensors, implementing advanced features like in-hardware motion detection, is crucial for computer vision in the IoT domain. Unfortunately, to be highly energy-efficient, these sensors might worsen the perception performance (e.g., resolution, frame rate, color). Therefore, domain-specific pipelines are usually delivered in order to exploit the full potential of these cameras. This paper presents the development, analysis, and embedded implementation of a realtime detection, classification and tracking pipeline able to exploit the full potential of background filtering Smart Vision Sensors (SVS). The power consumption obtained for the inference - which requires 8ms - is 7.5 mW.