Generalized Multi-camera Scene Reconstruction Using Graph Cuts

Abstract

Abstract. Reconstructing a 3-D scene from more than one camera is a classical problem in computer vision. One of the major sources of difficulty is the fact that not all scene elements are visible from all cameras. In the last few years, two promising approaches have been developed [11,12] that formulate the scene reconstruction problem in terms of energy minimization, and minimize the energy using graph cuts. These energy minimization approaches treat the input images symmetrically, handle visibility constraints correctly, and allow spatial smoothness to be enforced. However, these algorithm propose different problem formulations, and handle a limited class of smoothness terms. One algorithm [11] uses a problem formulation that is restricted to two-camera stereo, and imposes smoothness between a pair of cameras. The other algorithm [12] can handle an arbitrary number of cameras, but imposes smoothness only with respect to a single camera. In this paper we give a more general energy minimization formulation for the problem, which allows a larger class of spatial smoothness constraints. We show that our formulation includes both of the previous approaches as special cases, as well as permitting new energy functions. Experimental results on real data with ground truth are also included.

3D Reconstruction of Natural Underwater Scenes Using the Stereovision System IRIS

“…In [8], the energy minimization considers the input images symmetrically, handles visibility properly, and imposes spatial smoothness while preserving discontinuities. Fig.…”

“…Fig. 9 shows the dense disparity map obtained with the implementation of the graph-cut method presented in [8]. The disparity map gives us a dense correspondence map between the stereo images.…”

Plenoptic Layer-Based Modeling for Image Based Rendering

“…Even though the raw performance of our 1 st stage method is worse than that of [79], its disparity gradient estimates have a lower median error; this results in more accurate…”

“…Most segmentation methods of this type are based on spatial and colour similarities between pixels, two examples are the Mean Shift (MS) [73,78] and Graph Cut (GC) [79][80][81] algorithms.…”

“…3.2.5 improves the performance of both the alternative method (row 3) and our method (row 5). The disparity gradient histogram is here generated using either [79] or our 1 st stage method, the layers are assigned using the Lloyd-Max algorithm from Sec. 3.2.4 and the number of layers is 14 as indicated by the minimum sampling criterion, L min , from (2.10).…”