CenterNet: Keypoint Triplets for Object Detection

In object detection, keypoint-based approaches often suffer a large number of incorrect object bounding boxes, arguably due to the lack of an additional look into the cropped regions. This paper presents an efficient solution which explores the visual patterns within each cropped region with minimal costs. We build our framework upon a representative one-stage keypoint-based detector named CornerNet. Our approach, named CenterNet, detects each object as a triplet, rather than a pair, of keypoints, which improves both precision and recall. Accordingly, we design two customized modules named cascade corner pooling and center pooling, which play the roles of enriching information collected by both top-left and bottom-right corners and providing more recognizable information at the central regions, respectively. On the MS-COCO dataset, CenterNet achieves an AP of 47.0%, which outperforms all existing one-stage detectors by at least 4.9%. Meanwhile, with a faster inference speed, CenterNet demonstrates quite comparable performance to the top-ranked two-stage detectors. Code is available at

PDF Abstract ICCV 2019 PDF ICCV 2019 Abstract


Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Object Detection COCO minival CenterNet511 (Hourglass-52) box AP 41.3 # 149
AP50 59.2 # 85
AP75 43.9 # 72
APS 23.6 # 61
APM 43.8 # 59
APL 55.8 # 55
Object Detection COCO test-dev CenterNet511 (Hourglass-104, multi-scale) box mAP 47.0 # 119
AP50 64.5 # 89
AP75 50.7 # 78
APS 28.9 # 66
APM 49.9 # 64
APL 58.9 # 69
Hardware Burden None # 1
Operations per network pass None # 1