![]() ![]() Thus, instead of trying to improve the quality of the 360 ∘ on the foveal region, they reduce the resolution rendered in the peripheral region. However, the purpose of the foveation is to save the streaming bandwidth and computational power. Other researchers are attempting to take advantage of foveation phenomena in panoramic 360 ∘ video streaming. Their resulting images cannot be rendered directly for VR HMDs. However, these viewing approaches are impractical to implement for our application. Another way is to use a separate output display to render the output of each camera. Existing solutions to preserve the high-resolution detail is either to construct a 3D model of the scene or use a large high-resolution display. On the contrary, merging the low-resolution image onto the high-resolution one will require the upsampling of the low resolution, thus make the target rendered image bigger than the original. The foveal region on foveated images will not have the same high angular resolution to the PTZ camera, even if we try to digitally zoom in on the foveal region. However, merging the high-resolution image onto a low-resolution one will certainly reduce the detail captured in a stitched high-resolution region. Precisely, the common way in mosaicking different images are by performing an image stitching process in pixel level during graphics rendering. The problem of integrating the different resolution images is located on how we would preserve the high-resolution region detail. Several researchers have also tried to integrate the multi-resolution images that come from a single PTZ camera or a pair of wide-angle non-full panoramic 360 ∘ camera and PTZ camera for surveillance application. Moreover, we applied a zoom-based adjustment and blend mask for the seamless integration of the two images. Thus, the computational load of the modeling stage does not affect the frame rate of the foveated view. Regardless of the modeling stage, the rendering stage keeps updating frames from each camera. The positions of the primitives are updated only when the ROI (i.e., PTZ orientation) is changed. Instead of integrating two images in pixel level, we adjusted the positions of 3D geometric primitives in the modeling stage and rendered the camera images onto the corresponding primitives in the rendering stage. To do so, we separated our pipeline into two parallel stages: modeling and rendering. Unlike most foveated imaging research for wide FOV, we specifically designed our pipeline to support VR HMD, thereby maximizing the benefits of the use of omnidirectional camera-i.e., head synced view control and consequent increased spatial awareness. In this paper, we present a foveated imaging pipeline that integrates a wide-FOV image from an omnidirectional camera with a high-resolution ROI image from a PTZ camera. We discuss possible improvement points and future research directions. However, the improvement was less significant when the zoom level was 8× and more. Our evaluations showed that the proposed pipeline achieved, on average, 17 frames per second when rendering the foveated view on an HMD, and showed angular resolution improvement on the foveal region compared with the omnidirectional camera view. A control mechanism for placing the foveal region (i.e., high-resolution area) in the scene and zooming is also proposed. The pipeline consists of two parallel processes: one for estimating parameters for the integration of the two images and another for rendering images in real time. In this paper, we introduce a foveated imaging pipeline designed to support virtual reality head-mounted displays (HMDs). Conventional foveated imaging techniques may offer a solution for exploiting the benefits of both cameras, i.e., the high resolution of the PTZ camera and the wide field-of-view of the omnidirectional camera, but displaying the unified image on a 2D display would reduce the benefit of “omni-” directionality. ![]() Most cases use either PTZ or omnidirectional cameras exclusively even when used together, images from the two are shown separately on 2D displays. Pan-tilt-zoom (PTZ) and omnidirectional cameras serve as a video-mediated communication interface for telemedicine. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |