Yida Wang
Senior Research Engineer at Li Auto
Shenzhen, Guangdong Province, China
Summary
👤
Senior🎓
Top SchoolYida Wang is a Senior Research Engineer at Li Auto with 11 years of experience bridging academic research and production-grade computer vision systems. He holds a Ph.D. from the Technical University of Munich and has contributed research internships at Synthesia and Facebook focused on high-fidelity neural actors and 3D eye reconstruction for AR/VR. His open-source work includes notable contributions to OpenCV’s extra modules and the tiny-dnn C++ deep learning framework, where he implemented image generation for 3D models and advanced deconvolution/unpooling layers. Comfortable across C++ and ML pipelines, he combines low-level layer implementations with end-to-end feature extraction and backend integration. Based in Shenzhen, he currently leads efforts on a unified driving world model, applying 3D perception and learning techniques to production vehicle systems. A less obvious strength is his long-standing pattern of mentoring and open-source collaboration, from Google Summer of Code projects to enterprise research partnerships.
11 years of coding experience
2 years of employment as a software developer
Ph.D. Computer Science, Ph.D. Computer Science at Technical University of Munich
German, Japanese, English, Chinese
Github Skills (16)
neural-network10
mask-rcnn10
faster-rcnn10
computer-vision10
machine-learning10
deconvolution10
c-language10
convolutional-neural-networks10
caffe10
deep-learning10
cprogramming-language10
backpropagation10
image-generation9
algorithm9
algorithms9
Programming languages (5)
Github contributions (5)
tiny-dnn/tiny-dnn
Jun 2016 - Oct 2016
header only, dependency-free deep learning framework in C++14
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:29 commits, 14 PRs, 147 comments in 4 months
Contributions summary:Yida implemented deconvolutional and unpooling layers, demonstrating an understanding of convolutional neural network architectures. They contributed core code for forward and backward propagation within these layers, specifically targeting the efficient computation of these operations. The user's work directly aligns with the repository's focus on a deep learning framework, extending its capabilities for tasks involving deconvolution and unpooling operations. This included modifications for padding and the integration of a new backend style.
cppheaderdeep-learningc-plus-plusmachine-learning
opencv/opencv_contrib
Jun 2015 - Nov 2015
Repository for OpenCV's extra modules
Role in this project:ML Engineer
ML EngineerContributions:29 commits, 9 PRs, 14 comments in 4 months
Contributions summary:Yida implemented an image generator for 3D models, indicating a focus on computer vision or related areas. The commits include changes to `sphereview3d` module and modifications to `cnn_3dobj` including a feature extraction pipeline. This suggests the user was involved in generating and processing 3D object data, possibly for training machine learning models. The changes demonstrate integration with and use of Caffe.
pythoncmakecppcomputer-vision
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.