1,721,081 research outputs found
Interface design for character animation model
No full textThis report summarizes the research and development of the Final Year Project named as Interface Design for Character Animation Model. The team project was designed and developed by me together with my partners, Lim Pei Kin and Low Zhi Ling, under the supervision of Associate Professor Chau Lap Pui. The main objective of this project is to help the hearing impaired children to learn sign language more easily and in a fun and interactive way. For this project, my partners and I were assigned to create a program which can translate speech in English to Sign Language shown with 3D animation.Bachelor of Engineerin
Computer animation for hearing impaired application
No full textThis report documents the student’s 2 Semesters research and development of the Final Year Project (FYP) named Computer animation for hearing impaired application from year 2010/201l, under the supervision of Project supervisor Associate Professor Chau Lap Pui.
The idea of starting this FYP is to develop a fun and interactive application that will ease the hand sign language learning process of hearing impaired or deaf individual. It will also shorten the time require to pick up language and at the same time act as an education tools for those who wish to learn hand sign language.
The main purpose of this FYP is to enhance and fix errors encountered by previous batch FYP students. But due to the error is too hard to fix, a total new approach both in platform and software was used.Bachelor of Engineerin
Computer animation model for hearing impaired children
No full textThis report records the whole progress of my final year project with titled Computer Animation Model for Hearing Impaired Children, under supervision of Project supervisor Associate Professor Chau Lap Pui.
The main objective of this project is to develop a real-time sign-language learning tool for hearing impaired children through the 3D animated virtual character. The platform will convert speech or text input to animated sign language which is showed by a 3D character. This will be achieved through the creation of a computer graphic program which allows the hearing impaired children to interact with a cute and appealing virtual 3D character.
This whole project is developed by a team of three students; Lim Pei Kin, Thang Mei Nyok and me, each contributed in different part of project. This project is divided into 3 main parts; animation loader, speech recognition, program integration and user interface.
This project is referring to the previous batch student who did the similar application. However, the platform of our project is entirely new and different as compared to the previous project. The reason is because our main objective in this project is to develop the application in real-time platform by exploring the most powerful of C++ programming language.
My team and I were successfully completed this project in time. The basic requirements of the application are fulfilled. However, there are many improvements which can be done for future development.Bachelor of Engineerin
Human motion tracking with kinect sensors
No full textDeep learning in the field of object detection has been the upcoming topic of machine learning in the recent years. Resulting in the development of various deep learning frameworks and models to improve on the accuracy of object detection in different scenarios and application.
In this report, the author will be explaining the concept of computer vision and the increase in demand of computer vision and machine learning in multiple industries. Elaborating on the theory, limitations applications of deep learning methods in the area of computer vision.
This report provides a comprehensive study on a deep learning method that has been applied to a real-world scenario. The project is focused on identifying driver distraction. The author will demonstrate the process of gathering examples of driver distraction using Microsoft’s Kinect V2 to capture the actions. The author will then elaborate on the design of python scripts used to create the Convolutional Neural Network and the necessary processes that are needed to obtain a well-trained model.Bachelor of Engineering (Electrical and Electronic Engineering
Human motion tracking with kinect sensors
No full textDeep learning in the field of object detection has been the upcoming topic of machine learning in the recent years. Resulting in the development of various deep learning frameworks and models to improve on the accuracy of object detection in different scenarios and application.
In this report, the author will be explaining the concept of computer vision and the increase in demand of computer vision and machine learning in multiple industries. Elaborating on the theory, limitations applications of deep learning methods in the area of computer vision.
This report provides a comprehensive study on a deep learning method that has been applied to a real-world scenario. The project is focused on identifying driver distraction. The author will demonstrate the process of gathering examples of driver distraction using Microsoft’s Kinect V2 to capture the actions. The author will then elaborate on the design of python scripts used to create the Convolutional Neural Network and the necessary processes that are needed to obtain a well-trained model.Bachelor of Engineering (Electrical and Electronic Engineering
Motion detection for home security
No full textThis report serves to document the Final Year Project (FYP): Motion Sensing for Home
Security between Augusts 2015 and April 2016.
The purpose of this project is to develop Home Security Detection software that will rely on the web-camera as the device for image capturing. The program should be able to detect motion by the skeleton tracking technology. After motion is being detected, the program software will update the gesture quality index to show that how standard the motion is.
The therapists and the patient themselves can upload the data for analysis.
The author hope to supply a service that lets patients and their care providers view a digital version of their health records and improve the re-habitation service.
For this Motion Detection program, the author decided to use Visual Basic 2015 as her programming platform to code for this program. More details about the programs, programming and software techniques will be discussed in the report.Bachelor of Engineerin
3D graphics streaming over wireless communication network
No full textTechnology has evolved tremendously since the early 20th century. From personal computer to mobile phone, video and image streaming technologies are already well developed. However, the streaming technology for 3D graphics such as human motion capture (mocap) data is still far from practice use. Therefore, in this report, the author proposes ASF/AMC file format streaming to replace the existing 3D model or so called mesh data streaming based on JPEG 2000 streaming technique. For existing 3D model streaming, two changes which are automation reread process and increases in streaming rate, are made to enhance the overall performance of the system. Conclude that streaming based on JPEG 2000 Internet Protocol does not support well on 3D animation, author turn his attention to motion capture data. System analysis and design is craved and lead to the formation of system flow chart. Java is the programming language that is used throughout the development stages. The user interface of the client and server makes streaming request more robust.Bachelor of Engineerin
Pedestrian detection from surveillance camera
No full textAs the global cities and technology are growing quickly, the field about intelligent
monitoring has gradually become one of the main topics. As a major issue in the
field of intelligent monitoring, the problems of crowd counting gradually enter
people's field of vision. In crowded scenes, counting issues are very important for
safety and flow restrictions.
In recent years, convolutional neural networks (CNN) have achieved outstanding
results in the field of computer vision research. Its outstanding performance in image
feature extraction and model generalization effectively solves the feature extraction
problem of crowd counting under complex background. In view of the complexity of
some scenes, the current neural network models for crowd counting use deeper and
more complex structures to get the desired performance.
In order to get more efficient methods of extracting feature maps, in this report we
first analysis several typical models and their performance. Based on their drawbacks
then propose a regression-based neural network including residual net and two types
of attention module. The attention module is applied to both channel and spatial
dimensions, which can improve the feature extraction of the network model without
significantly increasing the amount of calculation and parameters. Besides, through
using various-scale architecture, we can also get high-resolution density maps. Then
simulation results on the dataset named ShanghaiTech and UCF_CC_50 show pretty
good performance compared to a few previous works.
At last, several actual problems and future research topics are presented in order to
make this model more practical.Master of Science (Communications Engineering
Image rendering for virtual reality application
No full textVirtual tour is more and more popular in recent years and implementation in reality is more and more frequent. Because it can easily link with other created graphics, images, animated films...so on. Furthermore virtually control the machine operation from remote location via communication channels. The objective of this final year project (FYP) is to research the function of 'Kinect' and how to utilize 'Kinect' in reality and virtual tour at home environment.
The report is categorized into five main sections: (1) Introduction, (2) Literature Review, (3) Design, Testing and Results, (4) Conclusion and Recommendations. The overall project description is described under "Introduction" section to give the readers about the overall project impression. Next, the programming and their functional and how it is being used. "Design and Simulation" section will describe start up and various testing on small skill programs which will be used in designing individual components. Then, followed by, "Results" section will describe the simulation and output of the project with the help of Visual Studio 2010 and C# software. Complying all the components and run the overall main program will show the project result in complete functionality of the 'Kinect'. After that, followed by "Conclusion and Recommendations' section for the conclusion of the project and future expansion and recommendations.Bachelor of Engineerin
Digital video broadcasting : terrestrial application development
No full textDigital Video Broadcasting (DVB) is defined as the internationally accepted standards for digital broadcast and has been the key solution to problems and limitations faced by analog broadcasting. DVB systems distribute data using a variety of methods most commonly of which include: satellite (DVB-S, DVB-S2), cable (DVB-C), terrestrial television (DVB-T) and terrestrial television for handhelds (DVB-H). With the tremendous advancement of DVB-T technology, the desire to develop DVB-T application on embedded system platform marks the new beginning of Digital Media era.
This project aims to develop a real time system to improve on live media streaming through a DVB-T tuner and the trimedia platform, via software decoding. It also involves the optimization of the software modules to improve the efficiency of decoding High Definition Televeision (HDTV). This is a continuation project which has been ongoing for two years, where the Philips Tuner will be replaced by a USB DVB-T Receiver, to be connected to the Nexperia Processor PNX 1500 to receive real time transport stream data and finally demultiplexed into audio and video streams.
This project has provided a very good learning experience for the student. It not only has allowed the student to gain a vast knowledge from the standards of MPEG 2. But also it has given a chance for the student to learn a lot more with regards to developing embedded system applications.Bachelor of Engineerin
- …
