Object Detection Techniques: A Comparison

• Published in: 2020 7th International Conference on Smart Structures and Systems (ICSSS) pp. 1 - 4
• Main Authors: Malhotra, Priyanka, Garg, Ekansh
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-07-2020
• Subjects: Classification algorithms; Computer vision; Conferences; Deep learning; Feature extraction; Machine learning; Measurement; Object detection
• DOI: 10.1109/ICSSS49621.2020.9202254

Computer vision is one of the technologies that aim at digitally perceiving the real world at a higher level through digital images and videos. Object detection, a subset to computer vision is one of the prominent techniques in this area of research. Object detection is basically an algorithm based on either machine learning or deep learning approaches employed for classification of elements in diverse classes and localization in the image. This paper provides a comparison among the three prominent approaches to achieve object detection. R-CNN, Fast R-CNN, YOLO are the techniques in the trend which facilitates the developer in accomplishing the task of detecting an object in the image. These techniques train and compute the parameters of the model in reduced hence increase performance as compared to the traditional object detection techniques.