I.1. Our first approach to high-performance imaging introduced nonfrontal imaging which uses a combination of sensor tilt and panning. It is implemented in the Omnifocus Camera which pans across any desired scene angle to obtain a seamless panoramic image with all objects captured in focus regardless of location, and, in addition, a range (depth) value at each pixel. Thus, the result is a registered pair of panoramic, omnifocused color and range images.

Additional desirable properties have been incorporated into the functioning of NICAM., such as: decreasing sensitivity to moving objects in achieving omnifocus, improving dynamic range, and extracting an entire surface in focus, instead of just one scene point as is usual.

Projects:

1. Imaging Scenes with Limited Motion using NICAM 2. Panoramic Imaging with Infinite Dynamic Range 3. Optimal Omnifocus Surfaces

I.2. One limitation of NICAM is that it pans and acquires images sequentially, and is thus not suited for scenes with fast moving objects. To overcome this problem, as well as for other applications, we have developed Split Aperture Imaging, and an associated multisensor Split Aperture camera. We have demonstrated the camera for the case of high dynamic range video imaging, over normal visual fields, using a novel optical beam-splitting and exposure control method.

I.3. To acquire panoramic video sequences, we have developed two types of Double-Mirror-Pyramid cameras that capture up to 360-degree fields of view at high-resolution. The first one, A Single View Double-Mirror-Pyramid Panoramic Camera, acquires a single sequence from one viewpoint, whereas the second, A Multiview Double-Mirror-Pyramid Panoramic Camera, provides multiple video sequences each taken from a different viewpoint, e.g. stereo sequences for 3D viewing. Both of these cameras belong to the family of pyramid cameras

I.4. To achieve a very large field of view, we have developed a Hemispherical Imaging Camera, which acquires high-resolution video sequences over a field of view that can range from being nearly hemispherical to being omni-directional, barring some small scene parts being obstructed by image sensors themselves.

Projects:

1. A Hemispherical Imaging Camera

I.5. We have developed a Table-top 3D camera that acquires the range image of an object, along with a registered color image at a resolution higher than that of the camera sensor. Together they provide a superresolution, 3D model of the object.

Projects:

1. A Single-Lens Depth Camera

I.6. In developing the new opto-geometric configurations, we have found that certain classical models and approaches cease to be adequate. For example, the long-established Gaussian model of image formation fails to adequately predict the acquired images, and the optical and geometric phenomena ignored in the traditional characterization of the most focused scene point make the traditional methods of focus analysis unacceptable. We have the old models with new, more rigorous, and satisfactory models. These new models are also useful in contexts other than next generation camera designs - they are useful in improving the performance of currently "acceptable" systems, and in extending the applicability of computer vision methods to many scenarios and applications which were out of reach otherwise.

Projects:

1. A New Imaging Model 2. A New Image Measure for Focus Evaluation

I.7. We describe a new omnidirectional stereo imaging system that uses a concave lens and a convex mirror to produce a stereo pair of images on the sensor of a conventional camera. The light incident from a scene point is split and directed to the camera in two parts. One part reaches camera directly after reflection from the convex mirror and forms a single-viewpoint omnidirectional image. The second part is formed by passing a subbeam of the reflected light from the mirror through a concave lens and forms a displaced single viewpoint image where the disparity depends on the depth of the scene point. A closed-form expression for depth is derived. Since the optical components used are simple and commercially available, the resulting system is compact and inexpensive. This, and the simplicity of the required image processing algorithms, make the proposed system attractive for real-time applications, such as autonomous navigation and object manipulation. The experimental prototype we have built is described.

Projects:

1. A New Omnidirectional Stereo Vision System Using a Single Camera

I.8. Related to the above new, high performance cameras is our earlier (1988-90) work on a new stereo vision system. We developed the University of Illinois Active Vision System to implement our active stereo vision algorithms for 3D vision.