HyperCam: Low-Power Onboard Computer Vision for IoT Cameras
This addresses the problem of deploying computer vision on resource-constrained IoT devices for applications like surveillance or monitoring, though it is incremental in improving efficiency for existing tasks.
The paper tackles enabling computer vision on low-power IoT cameras by proposing HyperCam, an energy-efficient image classification pipeline using hyperdimensional computing, achieving accuracies of 93.60% on MNIST and up to 72.79% on face identification tasks with low resource usage.
We present HyperCam, an energy-efficient image classification pipeline that enables computer vision tasks onboard low-power IoT camera systems. HyperCam leverages hyperdimensional computing to perform training and inference efficiently on low-power microcontrollers. We implement a low-power wireless camera platform using off-the-shelf hardware and demonstrate that HyperCam can achieve an accuracy of 93.60%, 84.06%, 92.98%, and 72.79% for MNIST, Fashion-MNIST, Face Detection, and Face Identification tasks, respectively, while significantly outperforming other classifiers in resource efficiency. Specifically, it delivers inference latency of 0.08-0.27s while using 42.91-63.00KB flash memory and 22.25KB RAM at peak. Among other machine learning classifiers such as SVM, xgBoost, MicroNets, MobileNetV3, and MCUNetV3, HyperCam is the only classifier that achieves competitive accuracy while maintaining competitive memory footprint and inference latency that meets the resource requirements of low-power camera systems.