My recent work of ResNeSt achieves superior performance on ImageNet and significantly boosts the performance on many downstream applications when serving as the backbone network. ResNeSt achieves state-of-the-art results in object detection and instance segmentation on MS-COCO dataset, and semantic segmentation on ADE20K and Cityscapes datasets.
I am also enthusiastic in contributing to open source projects. I am the designer and creator for AutoGluon Toolkit, PyTorch Encoding Toolkit. I am also a co-creator for GluonCV Toolkit and contribute frequently to Apache MXNet. I have organized “Everything you need to know to reproduce SoTA deep learning models” in ICCV 2019. I am organizing a tutorial series on “From HPO to NAS:
Automated Deep Learning” on CVPR 2020 and ECCV 2020.