Drive Safely Assistant for Android and JVM provides actionable information to a driver how to prevent a road accident. It recognizes driver’s drowsiness, distraction, belt presence, phone usage, eating, smoking, and camera sabotage.
Profile TensorFlow Lite model and measure its performance using FPS, model initialization time, inference time, and memory consumption on the smartphone. You can tweak model runs with different delegates (CPU, GPU, NNAPI, HEXAGON), XNNPACK option, number of threads.
The developed neural model solves an image classification task that predicts whether the human's eye is opened or closed using transfer learning technique.