Lingxiao He

算法研究员 at 美团

Beijing, China
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
Lingxiao He is an algorithm researcher with nine years of experience specializing in computer vision and person re-identification, currently based in Beijing and working at Meituan. He contributes actively to the well-known Fast-ReID repository, implementing and tuning Domain-Specific Re-identification (DSR) techniques that improve spatial feature reconstruction and re-ranking in evaluation pipelines. His work blends research rigor with production-minded engineering—integrating novel heads, evaluation modules, and distance metrics to boost re-id performance. Colleagues rely on him to translate algorithmic ideas into robust, tunable code that fits larger engineering systems. Notably, his hands-on commits show a focus on practical improvements (parameter tuning and pipeline integration) rather than only theoretical proposals.
code9 years of coding experience
github-logo-circle

Github Skills (5)

computer-vision10
pytorch10
python10
deep-learning9
algorithms8

Programming languages (1)

Python

Github contributions (5)

github-logo-circle
JDAI-CV/fast-reid

Jun 2020 - Sep 2020

SOTA Re-identification Methods and Toolbox
Role in this project:
userML Engineer
Contributions:26 commits, 1 PR, 25 pushes in 3 months
Contributions summary:Lingxiao primarily focuses on developing and refining DSR (Domain-Specific Re-identification) techniques within the Fast-ReID repository. Their commits demonstrate modifications to `dsr_head.py`, `dsr_evaluation.py`, and `dsr_distance.py` to implement, evaluate, and optimize the DSR algorithm. The changes involve integrating DSR into the evaluation pipeline and tuning parameters for improved re-identification performance. These contributions are closely related to spatial feature reconstruction and re-ranking methods.
pytorchre-rankingperson-reidbackbonetoolbox
JDAI-CV/Partial-Person-ReID

Sep 2017 - Sep 2020

Contributions:64 commits, 1 PR, 52 pushes in 3 years 1 month
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.
Request Free Trial
Lingxiao He - 算法研究员 at 美团