Kaiming He

E367295

Kaiming He is a prominent Chinese computer scientist known for pioneering deep learning architectures and techniques, including the influential ResNet model for image recognition.

All labels observed (2)

Label Occurrences
Kaiming He canonical 9
Kaiming 1

How this entity was disambiguated

Statements (48)

Predicate Object
instanceOf Chinese person ⓘ
computer scientist ⓘ
person ⓘ
researcher ⓘ
academicAdvisor Jian Sun ⓘ
awardReceived Best Paper Award at CVPR 2016 ⓘ
Marr Prize (ICCV) honorable mentions ⓘ
coAuthor Georgia Gkioxari ⓘ
Jian Sun ⓘ
Piotr Dollár ⓘ
Ross Girshick ⓘ
Shaoqing Ren ⓘ
Xiangyu Zhang ⓘ
educatedAt Chinese University of Hong Kong ⓘ
Tsinghua University ⓘ
employer Facebook AI Research ⓘ
linked to: Meta AI

Meta Platforms ⓘ
Microsoft Research Asia ⓘ
linked to: Microsoft Research
familyName He ⓘ
fieldOfWork artificial intelligence ⓘ
computer vision ⓘ
deep learning ⓘ
machine learning ⓘ
givenName Kaiming ⓘ
linked to: Kaiming He
knownFor Faster R-CNN ⓘ
linked to: FasterRCNN

Mask R-CNN ⓘ
linked to: MaskRCNN

MoCo (Momentum Contrast) framework ⓘ
R-CNN ⓘ
ResNet ⓘ
SimCLR-style contrastive learning variants ⓘ
deep residual learning ⓘ
deep residual networks for image recognition ⓘ
representation learning ⓘ
self-supervised learning methods ⓘ
spatial pyramid pooling in deep convolutional networks ⓘ
name Kaiming He ⓘ
nationality China ⓘ
notableContribution advances in object detection with region-based CNNs ⓘ
linked to: FasterRCNN

development of instance segmentation methods ⓘ
development of large-scale self-supervised visual representation learning methods ⓘ
introduction of residual connections in deep neural networks ⓘ
notableWork Deep Residual Learning for Image Recognition ⓘ
linked to: ResNet

Faster R-CNN (paper) ⓘ
linked to: FasterRCNN

Mask R-CNN (paper) ⓘ
linked to: MaskRCNN

Momentum Contrast for Unsupervised Visual Representation Learning ⓘ
Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition ⓘ
occupation computer scientist ⓘ
research scientist ⓘ

How these facts were elicited

Referenced by (10)

Full triples — surface form annotated when it differs from this entity's canonical label.

ResNet → developedBy → Kaiming He ⓘ
Kaiming He → name → Kaiming He ⓘ
Kaiming He → givenName → Kaiming ⓘ
linked to: Kaiming He
ResNeXt → paperAuthors → Kaiming He ⓘ
Shaoqing Ren → hasCoauthor → Kaiming He ⓘ
Jian Sun → coAuthorWith → Kaiming He ⓘ
Faster R-CNN → proposedBy → Kaiming He ⓘ
subject linked to: FasterRCNN
Mask R-CNN → introducedBy → Kaiming He ⓘ
subject linked to: MaskRCNN
RetinaNet → hasAuthor → Kaiming He ⓘ
Group Normalization → introducedBy → Kaiming He ⓘ