Detect people on a device that fits in the palm of your hands

This paper is about comparing different object detection models on a Raspberry Pi Zero 2 W using the TensorFlow Lite software. The models are filtered to only give the output of when it detects a person in the frame. Using different scenarios like walking into the room with a closed door, walking into the room with an open door, the object detection models got tested in different events. Each scenario was done once in light conditions and once in dark conditions to see if there is any difference in the models. All of the models that were chosen were models provider by TensorFlow themselves that were all trained with the COCO 2017 data set