Research Topics

 

 1. Computer Vison and 3D Visualization

Integral imaging is an autostereoscopic and multiscopic three-dimensional imaging technique that captures and reconstruction a light field by using a two-dimensional array of microlenses, sometimes called a fly's-eye lens, normally without the aid of a larger overall objective or viewing lens. In capture process, each microlens allows an image of the subject as seen from the viewpoint of that lens's location to be acquired. In reconstruction process, each microlens allows each observing eye to see only the area of the associated elemental image containing the portion of the subject that would have been visible through that space from that eye's location. The optical geometry can perhaps be visualized more easily by substituting pinholes for the microlenses, as has actually been done for some demonstrations and special applications. The result is a visual reconstruction complete with all significant depth cues, including parallax in all directions, perspective that changes with the position and distance of the observer, and, if the lenses are small enough and the images of sufficient quality, the cue of accommodation-the adjustments of eye focus required to clearly see objects at different distances. Unlike the voxels in a true volumetric display, the image points perceived through the microlens array are virtual and have only a subjective location in space, allowing a scene of infinite depth to be displayed without resorting to an auxiliary large magnifying lens or mirror.

此页面上的内容需要较新版本的 Adobe Flash Player。

获取 Adobe Flash Player


 2. Object Detection and Target Recognition

3D reconstruction from multiple images is the creation of three-dimensional models from a set of images. It is the reverse process of obtaining 2D images from 3D scenes. The essence of an image is a projection from a 3D scene onto a 2D plane, during which process the depth is lost. The 3D point corresponding to a specific image point is constrained to be on the line of sight. From a single image, it is impossible to determine which point on this line corresponds to the image point. If two images are available, then the position of a 3D point can be found as the intersection of the two projection rays. This process is referred to as triangulation. The key for this process is the relations between multiple views which convey the information that corresponding sets of points must contain some structure and that this structure is related to the poses and the calibration of the camera.

此页面上的内容需要较新版本的 Adobe Flash Player。

获取 Adobe Flash Player


 3. Digital Image Processing

Detecting objects in a 3D scene is an interesting challenge and can be useful for various computer-vision applications. For instance, detection of objects in 3D space has been proposed as a way to clearly present a scene in prosthetic vision devices. It is the intention that such devices will be used by blind people to aid their basic vision. Integral imaging techniques with which confocal images of depth planes are computationally reconstructed are potential inexpensive sources of rich 3D data.

Research Activities

2015.05.20 Prof. Piao, Invited Talk at HDHM NRC of South Korea

 

2014.11.13 Prof. Eun-Soo Kim, Director of HDHM visited DUT