官宣！恺明将于2024年入职MIT！恭喜！

Low-level

1. Single Image Haze Removal using Dark Channel Prior

   CVPR2009， Best Paper Award.

2. Guided Image Filtering

   ECCV2010, 图像降噪领域绕不开的边缘保持滤波器

3. Learning a Deep Convolutional Network for Image Super-Resolution

   三作，深度学习时代，图像超分开山之作

High-level Visual Recognition

1. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition

   ECCV2014，目标检测发展史上非常关键的一项工作。

2. Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification

   ICCV2015，深度学习方案首次在ImageNet上超越人类。

3. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

   二作，NeurIPS2015，目标检测发展史上关键基座之一。

4. Deep Residual Learning for Image Recognition

   CVPR2016，Best Paper Award

5. Identity Mappings in Deep Residual Networks

   ECCV2016，ResNetV2，影响力毋庸置疑了

6. Aggregated Residual Transformations for Deep Neural Networks

   五作，CVPR2017

7. Feature Pyramid Networks for Object Detection

   四作，CVPR2017，目标检测发展史上关键基座之一

8. Focal Loss for Dense Object Detection

   四作，ICCV2017，Best Student Paper Award

9. Mask R-CNN

   ICCV2017,Best Paper Award

10. Group Normalization

    二作，ECCV2018，Best Paper Honorable Mention

11. Panoptic Segmentation

    二作，CVPR2019，好像是首次开辟全景分割任务

12. Panoptic Feature Pyramid Networks

    三作，CVPR2019

Representation Learning

1. Momentum Contrast for Unsupervised Visual Representation Learning

   CVPR2020，Best Paper Nominee

2. Exploring Simple Siamese Representation Learning

   CVPR2021，Best Paper Honorable Mention

3. Masked Autoencoders Are Scalable Vision Learners

   CVPR2022，Best Paper Nominee，又一篇开创性的工作

以上仅列出何恺明一部分影响力甚大的工作，更多的相关工作可参考：https://kaiminghe.github.io/

最后，再次祝贺恺明即将加盟MIT，也期待未来更多具有有影响力的工作。

公众号后台回复“极市直播”获取100+期极市技术直播回放+PPT