Bin Yan
Research Scientist at ByteDance
San Jose, California, United States
Summary
🤩
Rockstar🎓
Top SchoolBin Yan is a research scientist based in San Jose with seven years of experience building generative AI and vision systems, currently working at ByteDance. He specializes in long-horizon interactive video generation and unified models for multi-task instance perception, with prior internships and collaborations at Google, ETH Zürich, Microsoft Research, and ByteDance that produced papers at WACV, CVPR, ECCV, ICCV. His work spans transformers for tracking (contributions to STARK) to real-time open-vocabulary video instance segmentation (TROY-VIS), emphasizing practical speed-accuracy trade-offs and data-driven processing pipelines. Bin holds a PhD in Electrical and Electronics Engineering and combines deep academic rigor with hands-on engineering, often adapting large foundation models (e.g., SAM) for tracking without labeled data. Notably, his projects balance cutting-edge research with production-minded efficiency, delivering models that are both fast and accurate for real-world deployment.
7 years of coding experience
Doctor of Philosophy - PhD, Electrical and Electronics Engineering, Doctor of Philosophy - PhD, Electrical and Electronics Engineering at Dalian University of Technology
English, Spanish
Github Skills (6)
transformer10
pytorch10
python10
data-processing9
machine-learning9
computer-vision9
Programming languages (4)
Github contributions (5)
researchmm/Stark
Mar 2021 - Mar 2022
[ICCV'21] Learning Spatio-Temporal Transformer for Visual Tracking
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:53 commits, 1 PR, 49 pushes in 1 year
Contributions summary:Bin's commits primarily focus on modifications within the `external/AR/ltr/data/processing.py` file, indicating involvement in data processing pipelines. These changes suggest the user worked on data transformations. The edits to `lib/train/scripts/base_functions.py`, `lib/models/stark/transformer.py`, and other modules, along with repository topics indicate a focus on transformer-based visual tracking.
pytorchiccvvisual-trackingspatio-temporaltemporal
MasterBin-IIAU/masterbin-iiau.github.io
Apr 2021 - Mar 2025
Contributions:49 pushes, 1 branch in 3 years 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.