Tianhe Ren
计算机视觉算法工程师 at International Digital Economy Academy 粤港澳大湾区数字经济研究院
Shenzhen, Guangdong Province, China
Summary
🤩
Rockstar🎓
Top SchoolTianhe Ren is a computer vision algorithm engineer with six years of experience focused on DETR-based detection and visual recognition, currently working at the International Digital Economy Academy in Shenzhen. Previously a senior computer vision engineer at IDEA-Research and an algorithm engineer at OneFlow, he combines research rigor from his PhD candidacy at CVMI-Lab with production-oriented model engineering. He has hands-on contributions to the popular detrex repository, tuning architectures, training pipelines, matchers and loss functions to support backbones like Swin and ConvNeXt. Tianhe’s background in software engineering from Xiamen University and his blend of academic and industry roles make him skilled at moving cutting-edge detection models from research into scalable implementations.
6 years of coding experience
工学士, Computer Software Engineering, 工学士, Computer Software Engineering at 厦门大学
Github Skills (12)
object-detection10
computer-vision10
pytorch10
custom-configuration10
configurations10
trainings10
loss10
system-configuration10
yml-configuration10
dtr10
python10
modeling10
Programming languages (7)
Github contributions (5)
IDEA-Research/detrex
Aug 2022 - Jan 2023
detrex is a research platform for DETR-based object detection, segmentation, pose estimation and other visual recognition tasks.
Role in this project:
ML Engineer
ML EngineerContributions:6 releases, 23 reviews, 84 commits in 5 months
Contributions summary:Tianhe contributed to the development and configuration of various models within the repository, specifically focusing on DETR-based object detection tasks. Their commits include aligning training configurations, refining model architectures and parameters, and adding support for different backbones and training setups, such as Swin and ConvNeXt backbones. The user also modified the loss functions and matchers used for model training, optimizing and adding new training scripts for the project.
eccvstate-of-the-articcvarxivdino
Oneflow-Inc/vision
Dec 2021 - Aug 2022
Datasets, Transforms and Models specific to Computer Vision
Contributions:165 reviews, 21 commits, 121 PRs in 8 months
imagenetvisionbackbonedeep-learningsemantic-segmentation
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.