Research Interest － Computer Vision and Deep Learning

​

• Image/video/3D point cloud understanding, segmentation, and detection

• Image editing and enhancement, e.g., image restoration, low-light enhancement, deblur and super-resolution

• Image/video/3D object/ 3D scene generation and reconstruction

• Multi-modal AI

• Other cutting-edge research in deep learning, e.g., semi- and self-supervised learning, knowledge distillation, few-shot learning, NAS, and AutoML

​

Open-source codebase

​

https://github.com/dvlab-research

​

​

Challenges

​​

• No. 1 of MEDIA AI Alibaba Challenge in Video Human Segmentation track 2020

• No. 1 of Apollo 3D Object Detection Challenge 2019

• No. 1 of Instance Segmentation Track in Scene Understanding Challenge for Autonomous Navigation in Unstructured Environments 2018 (PANet)

• No. 1 of WAD Drivable Area Segmentation Challenge 2018 (PSANet)

• No. 1 of LSUN Semantic Segmentation Challenge 2017 (PSPNet)

• No. 1 of COCO Detection Challenge in Instance Segmentation track 2017 (PANet)

• No. 2 of COCO Detection Challenge in Object Detection track 2017 (PANet)

• No. 1 of LSUN Instance Segmentation Challenge 2017 (PANet)

• No. 1 of ImageNet Scene Parsing Challenge 2016 (PSPNet)

• No. 3 of COCO Detection Challenge in Instance Segmentation track 2015