About Me
Hi, I’m Zhian Liu from South China University of Technolegy. Currently, I’m pursuing my master’s degree in computer science under the supervision of Yongwei Nie. I received my bachelor degree in software engineering from Wuhan University of Technology in 2020.
I am interested in programming, computer vision and some others cross-subject projects. And I’m curious about some low-layer techiniques in computer system, say, a DBMS. For my daily life, I’d like to do some reading about psychology and finance.
This site will contain what I learned, what I thought and what I experienced in my life.
Something I Did
Video Anomal Detection,[PDF] [Github]
Zhian Liu, Yongwei Nie, Chengjiang Long, Qing Zhang, and Guiqing Li.A HybridVideo Anomaly Detection Framework via Memory-Augmented Flow Recon-struction and Flow-Guided Frame Prediction.Proceedings of the IEEE Interna-tional Conference on Computer Vision (ICCV), Montreal, Canada, Oct 11-17, 2021
广告图片分类, [Github]
This project includes my solution of Ad. images classification. Basically, we train a classification model that leverages the image and text modalities together. We obtain 87.221% Top-1 accuracy on test set, which can meet practical application usage.
GrabCut plus Possion image blending,[Github]
This project is a combination of GrabCut and Possion image blending algorithms. We can use this tool to extract an object from an original image and blend it into another image. Implemented using C++ and OpenCV. The running time of this tool would be improved further some day.
SimpleDB: Chinese Version, [Github]
This project includes my Chinese translation scripts of the book Database Design and Implementation and reproduce code of SimpleDB, which is a toy DBMS written by Java.
