Binhan (Alex) Xu
Developer/Software Engineer
I am a software engineer at Pure Storage working on testing infrasturcture in Python. I graduated from Purdue University in December 2019 with a Bachelor of Science in Computer Engineering degree. I love working on cool projects and will be sharing those through by website.
- Location
- Mountain View, California, CA 94043, United States
- alex.xu1997@gmail.com
- Phone
- (765) 337-2421
- Website
- https://xu932.github.io
- Binhan (Alex) Xu
- GitHub
- xu932
Experience
– present
Software Enginner at Pure Storage
Pure Storage develops all-flash data storage hardware and software solution for various data centers.
Highlights
- Develop test infrastructure for new functionalities using pytest.
- Fix various bugs in software solution of Flash Array.
–
Software Engineer Intern - NBA 2K AI Team at Visual Concetps Entertainment
A video game company best known for sports game including NBA 2K series, NFL 2K series, etc. as well as other non-sports games such as Borderlands series and Civilization series.
Highlights
- Fixed various bugs/asserts from the bug database related to game play AI for NBA 2K20.
- Implemented new features to the passing system, the rebound system, and the double team system and improved game play.
–
Undergraduate Research Team Lead at Continuous Analysis of Many CAMeras (CAM2)
A research group that focuses on using images from network cameras to develop computer vision applications on embedded devices.
Highlights
- Improved the performance of the state-of-the-art computer vision algorithm You Only Look Once (YOLO) from 40 frames per second (FPS) to 260 FPS by utilizing parallelization and cache optimization on the supercomputing cluster.
- Exploited YOLO's inconsistency issue and developed a dataset consisting 3.72TB of images for researchers to develop more consistent object detectors.
Education
–
Bachelor in Computer Engineering
Purdue University
GPA 3.91
Courses
- Data Structure and Algorithm
- Object-Oriented Programming
- Microprocessor Systems and Interfacing
- Discrete Math
- Network Security
- Compiler
- Applied Algorithm (Graduate)
- Computational Models and Methods (Graduate)
- Machine Learning and Deep Learning (Graduate)
Publications
Large-Scale Object Detection of Images from Network Cameras in Variable Ambient Lighting Conditions by Caleb Tung, Matthew R. Kelleher, Ryan J. Schlueter, Binhan Xu, Yung-Hsiang Lu, George K. Thiruvathukal, Yen-Kuang Chen, Yang Lu
Computer vision relies on labeled datasets for training and evaluation in detecting and recognizing objects. The popular computer vision program, YOLO (“You Only Look Once”), has been shown to accurately detect objects in many major image datasets. However, the images found in those datasets, are independent of one another and cannot be used to test YOLO’s consistency at detecting the same object as its environment (e.g. ambient lighting) changes. This paper describes a novel effort to evaluate YOLO’s consistency for large-scale applications. It does so by working (a) at large scale and (b) by using consecutive images from a curated network of public video cameras deployed in a variety of real-world situations, including traffic intersections, national parks, shopping malls, university campuses, etc. We specifically examine YOLO’s ability to detect objects in different scenarios (e.g., daytime vs. night), leveraging the cameras’ ability to rapidly retrieve many successive images for evaluating detection consistency. Using our camera network and advanced computing resources (supercomputers), we analyzed more than 5 million images captured by 140 network cameras in 24 hours. Compared with labels marked by humans (considered as “ground truth”), YOLO struggles to consistently detect the same humans and cars as their positions change from one frame to the next; it also struggles to detect objects at night time. Our findings suggest that state-of-the art vision solutions should be trained by data from network camera with contextual information before they can be deployed in applications that demand high consistency on object detection.
Languages
- English
- Fluency: Native speaker
- Chinese
- Fluency: Native speaker
Skills
- Programming
- Level: ExpertKeywords:
- Version Control
- Level: ExpertKeywords:
- Machine Learing
- Level: IntermediateKeywords:
- Security
- Level: IntermediateKeywords:
- Parallel Computing
- Level: IntermediateKeywords:
- Web Development
- Level: BeginnerKeywords:
Interests
- LEGO
- Keywords:
- Sports
- Keywords:
- Games
- Keywords: