Comparison of YOLOv3 and SSD Algorithms
Ambika Neelopant, S. V. Viraktamath, Pratiksha Navalgi · Zenodo (CERN European Organization for Nuclear Research) · 2021
Object recognition is an advancement associated with computer vision and imaging, which manages to recognize and locate cases in computerized images and recordings for semantic artifacts of a particular class (such as persons, structures, or vehicles) in automated objects and observations. Continuous object identification and following is a huge, energetic yet uncertain and complex region of PC vision. It has end up being a noticeable module for various significant applications like video reconnaissance, self-sufficient driving, face identification; and so forth. As a feature of the overview, the theme investigated incorporate different calculations, quality measurements, speed/size trade-offs and preparing approaches. This paper centers around the two kinds of object identification YOLO (You Only Look Once) and SSD (Single Shot multi-box Detector) class of single step indicators and the Faster R-CNN class of two stage locators [1] and applications of the same.