Object detection is a computer vision (CV) task that identifies and localizes instances of predefined object classes in images, video, or point cloud frames.
The output is typically bounding boxes with class labels and confidence scores aligned to the project ontology.
Unlike classification, detection requires object location, and unlike segmentation, it does not require pixel-level outlines.
Common detector designs include two-stage models such as Faster R-CNN and one-stage models such as YOLO and SSD. Model performance is often evaluated with mean Average Precision (mAP) across Intersection over Union (IoU) thresholds.
The task is central to autonomous driving, robotics, security, smart home systems, and industrial inspection.
Object detection example in traffic scene
