Tianheng Cheng
Senior Researcher at ByteDance
Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolTianheng Cheng is a Senior Researcher in Beijing with a decade of experience advancing machine learning, deep learning, and visual perception, currently researching large multimodal models at ByteDance Seed. He holds a PhD in Artificial Intelligence from Huazhong University of Science and Technology and has a strong industry research background with internships at Tencent, Horizon Robotics, Microsoft Research Asia, and Face++. Tianheng has hands-on expertise in high-resolution visual recognition—contributing core implementations and refinements to the well-known HRNet facial landmark detection repository—demonstrating skill across model architecture, training, and inference pipelines. Known for bridging rigorous academic research with production-aware engineering, he navigates both foundational research and applied vision systems.
10 years of coding experience
Doctor of Philosophy - PhD, Artificial Intelligence, Doctor of Philosophy - PhD, Artificial Intelligence at Huazhong University of Science and TechnologyB.Eng, 电子信息工程, B.Eng, 电子信息工程 at 华中科技大学
Github Skills (8)
computer-vision10
pytorch10
model-validation10
trainings10
python10
face-alignment10
modeling10
inference9
Programming languages (5)
Github contributions (5)
HRNet/HRNet-Facial-Landmark-Detection
May 2019 - Jul 2019
This is an official implementation of facial landmark detection for our TPAMI paper "Deep High-Resolution Representation Learning for Visual Recognition". https://arxiv.org/abs/1908.07919
Role in this project:
ML Engineer
ML EngineerContributions:11 commits, 11 pushes, 25 comments in 2 months
Contributions summary:Tianheng primarily contributed to the implementation and refinement of the facial landmark detection model. Their commits include the initial setup of the HRNet model architecture, alongside subsequent modifications addressing issues with image flipping and transformations. They also made changes to core functions, demonstrating a focus on model training, validation, and inference pipelines within the repository.
representation-learningvisual-recognitionfacial-landmark-detectionrepresentationarxiv
wondervictor/DeepLearningWithPaddle
Mar 2017 - Mar 2018
Contributions:58 commits, 19 pushes, 1 branch in 11 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.