People

Zijian He

Computer Vision Engineer

I am a computer vision engineer in the mobile vision team. I joined Facebook at May 2015, worked on building computer vision system at Facebook scale from data collection tools to automatic training pipeline. In 2016, as mobile AI usage rose up, I started to focus more on optimizing computer vision algorithm for mobile device, including real-time style transfer, object classification, Mask-RCNN and depth prediction. I’m so excited to join the tech revolution of AI and impact people’s daily life.
 
Before joining Facebook, I graduated from UCLA as master in computer science. My advisor was Dr. Songchun Zhu. Thanks to his guidance, I rebuilt my mathematical understanding. I got my Bachelor degree at Zhejiang University, Hangzhou. It’s the most beautiful city in China where Alibaba located. I was born and raised in Guangzhou, the origin of Dim-sum. I enjoy taste different food in my spare time.

Interests

Image classification, object detection, neural network inference acceleration, single-image depth prediction, and neural network model optimization for embedded device.

Related Links

Website

Latest Publications

FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions

Alvin Wan, Xiaoliang Dai, Peizhao Zhang, Zijian He, Yuandong Tian, Saining Xie, Bichen Wu, Matthew Yu, Tao Xu, Kan Chen, Peter Vajda, Joseph E. Gonzalez

CVPR - June 1, 2020

Efficient Segmentation: Learning Downsampling Near Semantic Boundaries

Dmitrii Marin, Zijian He, Peter Vajda, Priyam Chatterjee, Sam Tsai, Fei Yang, Yuri Boykov

ICCV - August 12, 2019