Digital Fashion Show
Inspired by the visually impressive choreographed projection examples of the Puma LIFT ad campaign and various media events using Pepper's Ghost effects, we are developing simple calibration methods that support the projection of dynamic graphics content onto moving performers or other objects. This allows the designer to "paint" the actors with custom textures that may be responsive to their movements.
Stereo Acuity Test
CBC aired a 3D documentary Queen Elizabeth in 3D (QE3D) on 20/09/2010, which was the first 3D documentary to be broadcast in Canada. We developed a webpage where subjects can interactively measure their own stereo acuity, utilizing Randot® stereo test. It was linked in the CBC's QE3D page. [paper]
Medical Volume Carving
Segmentation of medical imaging data plays an important role in the medical community, with applications ranging from diagnosis to surgical planning, for example, defining the boundaries of a tumor, blood vessels, or the contours of the entire brain. In previous work, extensive effort has been invested in improving the performance of segmentation algorithms. However, comparatively little attention has been paid to the design of the human-computer interface for manual interaction with these systems. This includes the selection of input and display devices, but extends equally to the choice of interaction paradigm. [more] [video]
The World Opera
We conducted an experiment at Bang & Olufsen, comparing different display technologies (conventional video projection, Bang & Olufsen's 103" plasma display, and DNP's Holo screen) as suitable for stage performance. These displays were evaluated by a range of dancers interacting between two rooms. It is a part of the World Opera experiments. Key technologies include shadow removal, foreground/background segmentation, blob tracking, image blending, etc. [more]
Virtual Neurosurgical Biopsy
We developed a system to define a straight vessel-free path from the cortical surface to the targeted tumor using (a pen-like) probe. Because of the complex anatomical structure of the human brain and the topology of key blood vessels, defining such a path for insertion of probes and tools can be a challenging task in practice. In this system, we visualized the segmented brain vasculature with a simulated tumor at positions of varying difficulty of reaching. Users were provided with hand-coupled motion cues as they manipulated the volume and probe. [more]
3D Visualization and Gestural Interaction
This project deals with the challenges of medical image visualization, in particular within the domain of neurosurgery. We wish to provide an effective means of visualizing and interacting with data of the patient's brain, in a manner that is natural to surgeons, for training, planning, and surgical tasks. This entails three fundamental objectives: advanced scientific visualization, robust recognition of an easily learned and usable set of input gestures for navigation and control, and real-time communication of the data between multiple participants to permit effective understanding and interpretation of the contents. [more] [video]
Virtual Acupuncture
This project provides a comprehensive and accurate virtual human model for the study of Chinese acupuncture, based on the Chinese Visible Human Dataset. We achieve high quality rendering of a huge dataset by means of novel visualization and image compression technologies. The innovations of this project include efficient visualization of arbitrary cutting planes around each acupuncture point, accurate 3D positioning of more than 300 acupuncture points and 12 meridian systems, and force feedback to simulate needle puncture. [more]
Cartoon Deringing
Existing manga (or anime) are normally stored as JPEG (or MPEG). However, such BDCT encoding is not suitable for images with sharp edges, like manga/anime. Images are usually contaminated with annoying visual artifact. This paper presents an analogy-based deringing method to synthesize artifact-reduced images instead of traditional postfiltering approaches. Substantially visual and statistical improvements over existing methods are obtained.
[more] [paper] [software] [video]
[more] [paper] [software] [video]
GPU-Friendly Marching Cubes
Marching cubes has long been employed as a standard indirect volume rendering approach to extract isosurfaces from 3D volumetric data. This paper presents a GPU-friendly MC implementation. With the proposed parallel marching cubes algorithm, we can naturally generate layer-structured triangles, which facilitate the visibility-correct visualization of multiple-layer translucent isosurfaces without performing computational expensive sorting.