Appearance-based virtual view generation of temporally-varying events from multi-camera images in the 3D room

Abstract

In this paper, we present an "appearance-based" virtual view generation method for temporally-varying events taken by multiple cameras of the "3D Room", developed by our group. With this method, we can generate images from any virtual view point between two selected real views. The virtual appearance view generation method is based on simple interpolation between two selected views. The correspondence between the views are automatically generated from the multiple images by use of the volumetric model shape reconstruction framework. Since the correspondences are obtained by the recovered volumetric model, even occluded regions in the views can be correctly interpolated in the virtual view images. The virtual view image sequences are presented for demonstrating the performance of the virtual view image generation in the 3D Room. 9 cameras on the ceiling 10 cameras on each wall (40 in total)

Camera calibration using spheres: a semi-definite programming approach

“…Over the last few years, cameras have become ubiquitous due to their declining costs, so that vision systems using multiple cameras are becoming cost effective. Example of multi-view vision systems include the 3D room developed by Kanade et al [19] and the Keck Laboratory (Davis et al [9]). …”

Arbitrary Viewpoint Video Synthesis From Multiple Uncalibrated Cameras

“…Appearance-based virtual-view synthesis [16] takes into account the advantages of MBR and IBR. This 3-D shape model, which is recovered from multiple images, provides the required information for the IBR process, such as correspondence map, occluded area, etc., for the input images.…”

Estimation of Dense Displacement by Scale Invariant Polynomial Expansion of Heterogeneous Multi-View Images

“…Multi-view videos are widely used for 3-D visualization [1]- [3]. Multi-view images of a scene or an event are commonly captured by people holding different types of cameras.…”