• Rajesh Y R
• Aishwarya L
• Archana R G
• Brunda K O
• MD Irshad Hussain B

One of the common and difficult issues in computer vision is object detection. Researchers have widely experimented and contributed to the performance improvement of object detection and associated tasks including object classification, localization, and segmentation over the course of the last decade thanks to deep learning's rapid growth. Generally speaking, two stage and single stage object detectors are used to divide object detectors into two groups. Single stage detectors concentrate on all feasible spatial area suggestions for object detection via comparatively easier architecture in one go, whereas two stage detectors primarily focus on selected region proposals approach via complicated design. Any object detector's performance is assessed using its detection accuracy and inference time. In general, two stage object detectors outperform single stage object detectors in terms of detection accuracy. Single stage detectors have a faster inference time than their competitors, nevertheless. Additionally, the detection accuracy is increasing dramatically with the introduction of YOLO (You Only Look Once) and its architectural descendants, and occasionally it is better than two stage detectors. YOLOs are widely used in many applications, mostly because of their quicker conclusions rather than because of the accuracy of their detection. For instance, the detection accuracies for YOLO and Fast-RCNN are 63.4 and 70, respectively, whereas the inference time for YOLO is over 300 times quicker. We provide a thorough analysis of single stage object detectors, particularly YOLOs, regression formulation, developments in their design, and performance statistics in this work. Additionally, we include the applications based on two stage detectors, various variants of YOLOs, comparison illustrations between two stage and single stage object detectors, and future research prospects.

YOLO (You Only Look Once), ML (Machine Intelligence), AI (Artificial Intelligence), and OCR (Optical Character Recognition).

"YOLO BASED OBJECT DETECTION AND TRACKING USING AI & ML", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 7, page no.h203-h210, July-2023, Available :http://www.jetir.org/papers/JETIR2307724.pdf

"YOLO BASED OBJECT DETECTION AND TRACKING USING AI & ML", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 7, page no. pph203-h210, July-2023, Available at : http://www.jetir.org/papers/JETIR2307724.pdf