2015 · Computer vision
Deep Residual Learning for Image Recognition
Residual networks introduce skip connections that make very deep networks easier to optimize for visual recognition.
Paper & contextKaiming He
Making deep visual models easier to optimize and transfer.
He studies visual representations and the architectures used to learn them. These papers include residual networks, object detection, and segmentation, followed by work connecting representation learning with generative modeling.
Selected work
10 papers
2016 · Computer vision
Identity Mappings in Deep Residual Networks
This follow-up to ResNet studies why clean identity paths help very deep networks train.
Paper & context2015 · Computer vision
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Selected research in computer vision. Read the full paper, including the methods, experiments, and reported results.
Paper & context2017 · Computer vision
Mask R-CNN
Selected research in computer vision. Read the full paper, including the methods, experiments, and reported results.
Paper & context2021 · Computer vision
An Empirical Study of Training Self-Supervised Vision Transformers
Selected research in computer vision. Read the full paper, including the methods, experiments, and reported results.
Paper & context2026 · Computer vision
Video Generation Models are General-Purpose Vision Learners
Selected research in computer vision. Read the full paper, including the methods, experiments, and reported results.
Paper & context2026 · Computer vision
ELF: Embedded Language Flows
Selected research in computer vision. Read the full paper, including the methods, experiments, and reported results.
Paper & context2026 · Computer vision
Image Generators are Generalist Vision Learners
Selected research in computer vision. Read the full paper, including the methods, experiments, and reported results.
Paper & context2026 · Computer vision
GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Training
Selected research in computer vision. Read the full paper, including the methods, experiments, and reported results.
Paper & context2026 · Computer vision
Generative Modeling via Drifting
Selected research in computer vision. Read the full paper, including the methods, experiments, and reported results.
Paper & contextAn independent editorial profile. Inclusion does not imply Council membership or endorsement. Research is collaborative; coauthorship does not imply sole credit.
Selection & sources