DeepAI AI Chat
Log In Sign Up

CornerNet: Detecting Objects as Paired Keypoints

by   Hei Law, et al.
Princeton University

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 42.1


page 2

page 6

page 13


MetaAnchor: Learning to Detect Objects with Customized Anchors

We propose a novel and flexible anchor mechanism named MetaAnchor for ob...

Pair DETR: Contrastive Learning Speeds Up DETR Training

The DETR object detection approach applies the transformer encoder and d...

SiamCorners: Siamese Corner Networks for Visual Tracking

The current Siamese network based on region proposal network (RPN) has a...

FoveaBox: Beyond Anchor-based Object Detector

We present FoveaBox, an accurate, flexible and completely anchor-free fr...

Anchor Distance for 3D Multi-Object Distance Estimation from 2D Single Shot

Visual perception of the objects in a 3D environment is a key to success...

Oriented Objects as pairs of Middle Lines

The detection of oriented objects is frequently appeared in the field of...

Learning Fixation Point Strategy for Object Detection and Classification

We propose a novel recurrent attentional structure to localize and recog...

Code Repositories


This is a PPT for video of Follow_bobo AI workshop: CornetNet

view repo


An extension of the CornerNet architecture for RGB+T image inputs

view repo