YOLO-LITE: A Real-Time Object Detection Algorithm Optimized for Non-GPU Computers
This enables real-time object detection on portable devices like laptops and cellphones, though it is incremental based on YOLOV2.
The paper tackled real-time object detection on non-GPU devices by developing YOLO-LITE, achieving 21 FPS on a non-GPU computer and 12.26% mAP on COCO.
This paper focuses on YOLO-LITE, a real-time object detection model developed to run on portable devices such as a laptop or cellphone lacking a Graphics Processing Unit (GPU). The model was first trained on the PASCAL VOC dataset then on the COCO dataset, achieving a mAP of 33.81% and 12.26% respectively. YOLO-LITE runs at about 21 FPS on a non-GPU computer and 10 FPS after implemented onto a website with only 7 layers and 482 million FLOPS. This speed is 3.8x faster than the fastest state of art model, SSD MobilenetvI. Based on the original object detection algorithm YOLOV2, YOLO- LITE was designed to create a smaller, faster, and more efficient model increasing the accessibility of real-time object detection to a variety of devices.