I am a Researcher at ByteDance (Toutiao) AI Lab in Menlo Park, CA. Before that, I was a Senior Research Scientist at Adobe Research. I obtained my Ph.D. degree in the Department of EECS at Northwestern University in June 2013, advised by Prof. Ying Wu. I previously had summer internships at Adobe in 2012 and 2011, and at Nokia Research Center (Santa Monica, CA) in 2010. I received my B.S. and M.S. degrees from the Department of Automation at Tsinghua University, China.
I am generally interested in the research topics in the area of Computer Vision, particularly semantic segmentation, depth prediction, object detection and recognition, image editing and deep learning-related problems.
Email: shenxiaohui [at] bytedance [dot] com | Google Scholar | LinkedIn
News
- Our research team in Menlo Park is looking for researchers, engineers and interns in the area of Computer Vision, Graphics and Machine Learning all year round. Feel free to contact me if you are interested.
- One paper on Single Image Shallow Depth-of-field was conditionally accepted by SIGGRAPH Asia.
- Four papers were accepted by ECCV 2018.
Selected Projects
Semantic-aware Image Editing
- Deep Image Harmonization
CVPR 2017 | [Project Page] [Paper] [Model/code] [Try it in Photoshop Mix] - Sky is Not the Limit: Semantic-Aware Sky Replacement
SIGGRAPH 2016 | [Project Page] [Adobe MAX Demo]
Semantic Segmentation
- Human Body Parsing/Object Parsing, Collaboration with SYSU/NUS/UCLA
[Look Into Person: Human Parsing and Pose Estimation Challenge]
CVPR 2017 | ECCV 2016 | CVPR 2016 | ICCV 2015a | ICCV 2015b | CVPR 2015 - Scene Parsing, Collaboration with UC Merced/NUS
Image Scene Parsing, ICCV 2017 | Urban Scene Parsing, ICCV 2017 | Video Scene Parsing, ICCV 2017
Image Quality and Aesthetics
- Shipped in Aesthetic Filters in Adobe Stock, Auto Curate in Elements Organizer 2018, Candid Moments in Premiere Elements 2018
- Personalized Image Aesthetics
ICCV 2017 | [Paper] [Dataset] - Photo Aesthetics Ranking Network with Attributes and Content Adaptation
ECCV 2016 | [Paper] [Dataset] - Deep Multi-Patch Aggregation Network for Image Style, Aesthetics, and Quality Estimation
ICCV 2015 | [Paper]
