Efficient 3D thing Segmentation from Densely Sampled Light Fields with Applications to 3D ReconstructionComputer Graphics research from ETH Zurich and Disney research demonstrates method to improve 3D thing reconstruction through photogrammetry method,and includes some spicy visuals on how the process works:
Precise thing segmentation in image data is a fundamental problem with various applications, including 3D thing reconstruction. We present an efficient algorithm to automatically segment a static foreground thing from highly cluttered background in light fields. A key insight and contribution of our paper is that a meaningful increase of the available input data can enable the design of novel, or highly efficient approaches. In specific,the central idea of our method is to exploit high spatio-angular sampling on the order of thousands of input frames, e.g. captured as a hand-held video, and such that new structures are revealed due to the increased coherence in the data. We first show how purely local gradient information contained in slices of such a dense light field can be combined with information approximately the camera trajectory
to execute efficient estimates of the foreground and background. These estimates are then propagated to textureless regions using edge-aware filtering in the epipolar volume. Finally,we enforce global consistency in a gathering step to derive a precise thing segmentation both in 2D and 3D space, which captures fine geometric details even in very cluttered scenes. The design of each of these steps is motivated by efficiency and scalability, and allowing us to handle large,real-world video datasets on a standard desktop computer. We demonstrate how the results of our method can be used for considerably improving the speed and quality of image-based 3D reconstruction algorithms, and we compare
our results to state-of-the-art segmentation and multi-view stereo methods. More Here
Source: tumblr.com