I am a graduate student at Beihang University (BUAA), majoring in Mechanical Engineering and minor in Subject Teaching (Phisics). I earned my bachelor degree from University of Science and Technology Beijing (USTB), majoring in Automation Engineering and supervised by Prof. Ling Wang. I had a research experience working on Data Mining, Augmented Reality and Computer Vision. I had a few academic exchange experiences abroad, in Queensland University (UQ, Australia) in 2018 and National University of Science and Technology (NUST MISIS, Russia) in 2019.

My research interests lie in Computer Vision and Pattern Recognition. Currently I am engaged in reseaching NeRF & 3DGS, 6D Pose Estimation, Augmented Reality and Modern Education Technology.

At present, I am an intern teacher of Senior Physics I in Beijing National Day School (BNDS) and preparing for the Tin Ka Ping Cup.

Research

NeRF & 3DGS

3D Reconstruction based on Instant-NGP(NeRF)

3D Reconstruction based on 3DGS

6D Pose Estimation

6D Pose Estimation using iNeRF

MP6D industrial objects Pose Estimation using PVN3D

Assembly Guidance System based on Augmented reality (Hololens)

[ Video ] [ PDF ]

In this research, An AR aided assembly system is designed and implemented. The database on the server side is mainly built through MySQL8.0, which is used for client to call and modify user information and process data. Applications on HoloLens are mainly built through Unity3D 2019.3LTS. Functions such as human-computer interaction, device communication and assembly guidance are realized through C# scripts. Image recognition and 3D registration are realized through Vuforia or ARToolKit. The application on the PC server is mainly built by Visual Studio 2019, and the interactive function, database call, device communication and other functions are realized by C# code. The interaction between HoloLens and the server mainly considers the problem of HoloLens’ computing power. It adopts the scheme of separating partial image classification and recognition from HoloLens devices, puts the image processing and recognition module in HoloLens on the server, and transmits the pictures taken by HoloLens to the server through TCP protocol. HoloLens voice module is also mainly used to receive user commands and send commands to the server.

Air pollution region division based on deep belief network

Automatic dispensing of spacecraft heating sheet based on machine vision

Defect detection methods for highly reflective surfaces

Augmented reality assisted blind operation in small Spaces (HoloLens)

Straight line detection methods in cluttered scene