1Hardgeneralโฑ 45sWhy do modern object detection architectures like Faster R-CNN use a Region Proposal Network (RPN)?๐โผATo classify the entire image as a single labelBTo efficiently generate candidate bounding box regions likely to contain objects, avoiding the need to exhaustively evaluate every possible region at every scaleCTo perform only image segmentation, not detectionDRPNs are only used for facial recognition