Other Projects

Here you can find other interesting projects I have done, either for a course or just for fun!! =)

Application for Parallel Computing

Project Idea

Augmented Reality

Computer Vision, Computational Photography, Arts

snapIn this project, we built an Android App that achieves augmented reality. More specifically, we are able to place a user-selected food image onto a plate that locates in a real-world scene. We used “FAST” feature for matching and developed the algorithm with the OpenCV library. We also made a Youtube video on our project. Here is the link.


3D Pokemon-version Snake Game with OpenGL

Computer Graphics, Arts


3D village scene

My first time with OpenGL and 3D scene rendering! It’s a simple snake game, but with the ‘food’ being Pikachu and the ‘snake’ being a Pokeball; and the environment is a fancy village with houses, trees, a cart and a sun that changes lighting of the scene. Our Pokeball increases one when encountering a Pikachu, and decreases one when encountering a Miaouss. The Pokeball dies whenever it hits a house or a tree.

I have learned a lot from this project about using the OpenGL library with C++ and basic graphics knowledge. I will talk about some technical details about this game development in my blogs.

Classifier Comparison for Face Recognition

Computer Vision, Machine Learning

 Face datasetThis project evaluates different classifiers’ performances in face recognition tasks. The dataset poses challenges for face recognition as the intra-class variances aren’t very significant compared to the within-class variances in the dataset. However, this is often the case with real-life face recognition tasks. Another thing we are concerned about is the non-sufficient training samples for each class, as each input vector is of high dimension.

The report of this project can be viewed here

Image Classification with Feature Descriptors

Computer Vision, Signal Processing

Images are trained by searching for feature descriptors and building up clusters that could best represent particular classes. By matching feature vectors of each test image with the cluster of the training dataset, test images are classified. Different kinds of feature descriptors such as SIFT, PCA-SIFT and Vocabulary Tree are implemented and compared.

The report of this project can be viewed here

Virtual Drum Kit

Computer Vision, Signal Processing

Our design features amateurs who want to experience a cool drum kit but who also desire to save money with a cheaper option. The system can be divided into three major components: video processing, which tracks the current position of drum stick in real time; audio processing, which detects the ‘hit’ of the drum stick and generates drum sounds; and an integration of the previous 2 parts.

The poster of this project can be viewed here

Binocular Stereo

Computer Vision

depth Retrieve depth information from multiple images and construct a disparity map of the scene.

The report of this project can be viewed here

Performance of Indoor MU-MIMO with mobility

Network Communication


Multi-User MIMO has a direct theoretical capacity gain that’s proportional to the number of nodes, which are not required to be multi-antenna terminals. This leads to high flexibility in terminal size and is beneficial for small mobile devices such as smartphones, tablets and laptops that have too small infrastructure to include multiple antennas.

However, we noticed that these portable small devices, especially smartphones and tablets are usually attached with high mobility when in use. We think that mobility should be taken into account for affecting throughput in MU-MIMO communication.

The paper of this project can be viewed here (source code link is included in the paper)


Embedded System


In my junior year I took a course in embedded system where we hacked a crazyflie that employs a 32-bit microcontroller unit (MCU) from ST Microelectronics based on ARM’s Cortex M3 architecture. The majority of development takes place in this MCU.

I created a web page along with this project, which can be viewed @ Project Web Link (related source codes are attached)



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s