Back to Search
Start Over
CornerNet: Detecting Objects as Paired Keypoints
- Publication Year :
- 2018
-
Abstract
- 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.<br />Extended version with additional results. Test AP on MS COOO improved from 42.1% to 42.2% after a bug fix
- Subjects :
- FOS: Computer and information sciences
Computer science
business.industry
Computer Vision and Pattern Recognition (cs.CV)
Pooling
Detector
Computer Science - Computer Vision and Pattern Recognition
Pattern recognition
02 engineering and technology
Object (computer science)
Convolutional neural network
Object detection
Set (abstract data type)
Artificial Intelligence
Minimum bounding box
Pattern recognition (psychology)
0202 electrical engineering, electronic engineering, information engineering
020201 artificial intelligence & image processing
Computer Vision and Pattern Recognition
Artificial intelligence
business
Software
Subjects
Details
- Language :
- English
- Database :
- OpenAIRE
- Accession number :
- edsair.doi.dedup.....96b2dd7178cd66fb7e46b262a6465ceb