作者单位:University of Michigan, Ann Arbor
发布时间:2018
发布期刊/会议:ECCV
**论文全称:**CornerNet: Detecting Objects as Paired Keypoints
论文地址:https://arxiv.org/abs/1808.01244
论文代码:https://github.com/umich-vl/CornerNet
地位:无需锚框的物体检测算法的开山鼻祖
We propose CornerNet, a new approach to object detection where we detect 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. In addition to our novel formulation, we introduce corner pooling, a new type of pooling layer that helps the network better localize corners. Experiments show that CornerNet achieves a 42.2% AP on MS COCO, outperforming all existing one-stage detectors.