Final Year Student Projects: Prelude Michael R. Lyu

Video over InternEt and Wireless (VIEW) Technologies  Video  Internet  Wireless

Wireless Video-On-Demand Internet is ubiquitous and free Video is key to rich media content Wireless services are chargeable

Wireless News-On-Demand Stay in touch Redeem unused time Quick search Portable information Continuous flow Dynamic delivery

Play Interactive Wireless Videos

Recreation Shopping Interaction Education Trading Application Areas Entertainment

Wireless-Enhanced Digital Video Content Hub

Final Year Student Projects: Project Listing Michael R. Lyu

Background: This is a project that students can select to implement a wireless PDA Application or Game. Platforms & Technologies: PDA platform: PocketPC or Palm devices Wireless environment: Bluetooth, b, or GPRS (so- called 2.5G) Attempt on bluetooth is encouraged. Programming languages: Embedded Visual C++ (Pocket PC), native C (Palm), J2ME (mobile phones) Project One: Innovative Wireless PDA Application/Game

Suggested Applications: 1. Multiplayer PDA-PDA Game over Bluetooth There is no such kind of game in existence yet. We are ordering the latest SDK for iPAQ H3870 development. Successfully implementation of a complete game will have the demonstration opportunities to local companies. 2. Interactive Presentation/Control with PDAs Enhancing the current wireless PDA presentation tool for supporting multiple users and computer agents interactions. Use PDA as a generic control user interface for miscellaneous toys and devices. Project One: Wireless PDA Application/Game

Suggested Applications: 3. Payment System for Wireless Access An application with high commercial value. To develop a desktop Linux based system to redirect clients to pay money before they can make use of the wireless access service. This application can be integrated with Application J2ME Based Personal Information Management System Develop a wireless and web based framework that allows users to access and manipulate personal information anytime and anywhere. It targets for deployment in the growing J2ME handset population and 2.5G environment. A team coming up with a sound application will have opportunities to demonstrate that to Sun Microsystems or enter a competition. Project One: Wireless PDA Application/Game

Requirement: Basic hardware knowledge. High EQ to face frustrating situations. Sensitive to the technology trend. Nevertheless, hardworking and creativity. Project One: Wireless PDA Application/Game

Project Two: Advanced Audio Information Retrieval System Background: As more and more audio data is captured and stored on computers and the Internet, effective automatic audio indexing and retrieval systems are needed in order to fully use audio information for index and searching. Audio classification and segmentation are the pre-requisites to automatic indexing and retrieval.

Project Two: Audio Information Retrieval System Audio Signal Feature Extraction Speech Non-speech Male? Female? Speaker Identification/ Classification Music Pattern Matching Audio Scene Change From News’ Audio Channel MFCCSegmentation Detect cont’ vowel > 30% By MFCC var. By 256 GMM By Clustering

Project Two: Audio Information Retrieval System Project Description: In this project, students will investigate the characteristics of audio, extract useful information, perform audio indexing and classification. It will include music/speech discrimination, audio scene change and speaker identification, and music pattern matching. Techniques: Signal Processing, Pattern Recognition, Matlab and Visual C++

Project Three: Smart Traveller with Visual Translator In this project, students will combine digital camera, Personal Data Assistant (PDA), Optical Character Recognition (OCR), and even Face Recognition techniques to form a visual translator for travellers (VTT). Signs written in foreign characters are hard to understand. Travellers will use the VTT to take the pictures of the sign and mark the region of the target foreign characters. VTT then will recognize the foreign text and translate it into native language. Possible (and ambitious) extension can include face recognition engines to remind the travellers about people’s names.

Project Three: Smart Traveller with OCR “Footbridge”

Project Three: Smart Traveller with Face Recognizer “This Guy is Michael”

Project Three: Smart Travellers Techniques: PDA programming, Image processing, OCR Text Searching Face recognition Platform: Palm OS Summer job will be offered to develop an OCR engine for on-screen subtitle in TV broadcast.

Project Four: Augmented-Reality for Video Content In this project, students will apply augmented- reality techniques which dynamically superimposes interactive computer graphics images onto objects in the real world videos. The target is to create photo realistic graphics which match exactly with the environment lighting characteristics. Using the above techniques to construct augmented-reality virtual sets for film industry

Project Four: Augmented-Reality for Video Content

Techniques: Video processing Computer vision and pattern recognition Computer graphics, realistic photo generation Platform: Win32

My previous FYP students Justin Wan: ACM Team Kenny Kwok: ACM Team Starsky Wong: ACM Team Ho Man Lam: ACM Team Yeung Kam Wah: ACM Team Lam Yee: ACM Team

My Previous FYP Students