About AMP Lab Projects Downloads Publications People Links
Howard
Wing Ho Leung Personal Homepage:
http://www.cs.cityu.edu.hk/~howard/ |
||
Office: Porter Hall B43 Lab: Porter Hall B6 Phone: 412-268-7100 Fax: 412-268-1679 |
Mailing Address: Porter Hall B6, Department of ECE, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213-3890, USA |
[Research Interests] [Project] [Publications] [Joint Collaboration between CMU and IBM]
Sketch Retrieval
Using a pen-based device as input provides the user a more natural way of interaction compared with other input devices such as the keyboard. As a result, this motivation opens up a new field of research for improving pen-computing technologies in order to justify the use of these pen-based devices. For example, in the classroom, the lecture notes written by the teacher on the whiteboard can be captured electronically by a collaboration device such as "mimio". Later students can retrieve relevant lecture materials from the hand-drawn sketch database by sketching a query drawing. Each sketch (hand-drawing) in the database consists of strokes that are sequences of coordinates of the points sampled by the pen-based device. Retrieval in the hand-drawn sketch database is equivalent to finding a stroke or multiple strokes from the database that are a good match to the query stroke(s).
Image-Based Rendering (IBR)
For Image-Based Rendering applications, different views of a scene can be reconstructed by processing the images acquired at some fixed views of that scene. Often the number of captured images is very large, we have to store these images efficiently. In other words, can we compress these images in such a way that we obtain the most efficient representation of these IBR images ?
3D Virtual Conferencing Environment
In a virtual 3D conferencing environment, users can communicate with each other freely by using animated human-like characters known as avatars to represent themselves. We can apply various audio-visual communication technologies such as real-time spatialized audio or face and eye tracking to provide consistent spatial relationship among users, both visually and acoustically, and to preserve the eye contact and the sense of immersion. Moreover, we can use texture mapping or image-based rendering to provide a more realistic look of an object or the environment in this virtual space.
NetICE When multiple participants are involved in video conferencing, current software places different participants in different windows. This arrangement makes it difficult for the participants to tell who is talking to whom, and there is no sense of immersion. By enhancing the capabilities of the network and adding intelligence to the end terminal, the quality of the audio-visual communication can be improved. In this project, we would like to provide a system that will reproduce an actual conference setting, so that people can communicate from anywhere at any time and still feel that they are in the same environment without realizing the limitation of the physical network. For more details on the project please see the NetICE project page. |
|
Hand-Drawn Sketch Retrieval In this system, the user is able to draw a query and the system will search through similar sketches from a collection of free-form hand-drawings. For example, in the classroom, the lecture notes written by the teacher on the whiteboard can be captured electronically by a collaboration device such as "mimio". Later students can retrieve relevant lecture materials from the hand-drawn sketch database by sketching a query drawing. For more details on the project please see the Hand-Drawn Sketch Retrieval project page.
|
|
Trademark Retrieval In traditional trademark retrieval systems, the trademarks are first annotated with keywords that are then used for retrieval. However, this process requires a lot of manual labor in order to assign keywords. Also, the annotation for the same trademark may not be consistent with different users. As a result, it is more desirable to allow the user to input a query by providing a rough sketch and then the system is smart enough to automatically extract features from this query sketch and use them for comparison in order to search for similar trademarks in the database. For more details on the project please see the Trademark Retrieval project page. |
Journal:
Conference:
Technical Report and White Paper:
Thesis:
This
website is maintained by
Devi Parikh
|