Moving beyond image classification, this section addresses the challenge of locating and identifying multiple objects within an image. This is the objective of object detection. Here, you will study the design and operation of influential deep learning based object detection models.

We will analyze the architectural differences between two main approaches: two stage detectors, exemplified by the R CNN series (including Faster R CNN with its Region Proposal Networks), and computationally faster single stage detectors like YOLO and SSD. You will gain practical understanding of essential components such as anchor boxes used for proposing potential object locations and techniques like non maximum suppression ($NMS$) for filtering overlapping detections. We will also cover standard evaluation protocols, focusing on metrics like mean Average Precision ($mAP$), and conclude with hands on implementation practice.