Next Generation Cameras
Our work in this area is aimed at defining technologies for the next generation cameras. Traditional digital cameras suffer from many limitations, including limited: field of view, depth of field, dynamic range, frame rate, and resolution; and the image acquired may not be associated with a single viewpoint. The early applications of digital cameras were straight-forward, and not seriously affected by these limitations. However, with increasing sophistication of machines, communication and users, the diversity and complexity of tasks implemented on digital cameras are steadily growing, and the above limitations are becoming increasingly restrictive. While a single application may not be simultaneously affected by all the limitations, research on next generation cameras must address all of them, so as to match user expectations and spur new applications.
Since early 1990’s, we have been engaged in introducing new technologies to overcome various of the aforementioned limitations, and developing associated high-performance camera systems. A central feature of our approaches is that they integrate principles from optics, signal processing, electronics, computer and natural vision, computer algorithms and mechanical design, to harness their combined strengths. The integration leads to capabilities often infeasible to achieve within the current state of the art in the individual disciplines. Such approaches have formed the new field of high-performance cameras which has attracted a large number of researchers in academia and industry worldwide since late 1990’s.
The projects below summarize the camera technology developed in the lab. All of them possess the single viewpoint property but offer different trade-offs among the field of view, depth of field, dynamic range, frame rate, and resolution.