Description
CornerNet is an object detection model that detects an object bounding box as a pair of keypoints, the top-left corner and the bottom-right corner, using a single convolution neural network. By detecting objects as paired keypoints, we eliminate the need for designing a set of anchor boxes commonly used in prior single-stage detectors. It also utilises corner pooling, a new type of pooling layer than helps the network better localize corners.
Papers Using This Method
Robust Table Detection and Structure Recognition from Heterogeneous Document Images2022-03-17TetraPackNet: Four-Corner-Based Object Detection in Logistics Use-Cases2021-04-19RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder2020-10-29MatrixNets: A New Scale and Aspect Ratio Aware Architecture for Object Detection2020-01-09CenterNet: Keypoint Triplets for Object Detection2019-04-17Bottom-up Object Detection by Grouping Extreme and Center Points2019-01-23CornerNet: Detecting Objects as Paired Keypoints2018-08-03