Conference Paper

Quasi-Dense 3D Reconstruction using Tensor-based Multiview Stereo

DOI: 10.1109/CVPR.2010.5539796 Conference: The Twenty-Third IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2010, San Francisco, CA, USA, 13-18 June 2010
Source: DBLP


We propose tensor-based multiview stereo (TMVS) for quasi-dense D reconstruction from uncalibrated images. Our work is inspired by the patch-based multiview stereo (PMVS), a state-of-the-art technique in multiview stereo reconstruction. The effectiveness of PMVS is attributed to the use of 3D patches in the match-propagate-filter MVS pipeline. Our key observation is: PMVS has not fully utilized the valuable 3D geometric cue available in 3D patches which are oriented points. This paper combines the complementary advantages of photoconsistency, visibility and geometric consistency enforcement in MVS via the use of 3D tensors, where our closed-form solution to tensor voting provides a unified approach to implement the match-propagate-filter pipeline. Using PMVS as the implementation backbone where TMVS is built, we provide qualitative and quantitative evaluation to demonstrate how TMVS significantly improve the MVS pipeline.

1 Follower
44 Reads
  • Source
    • "However, the improvement is determined by the number of available processing units. More recently, [11] and [12] proposed a different weighting factor to be used instead of (9) which aims at avoiding its discontinuity. The introduction of this weighting factor simplifies the computations, but at a cost of yielding very different values from those obtained through the original tensor voting. "
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes two alternative formulations to reduce the high computational complexity of tensor voting, a robust perceptual grouping technique used to extract salient information from noisy data. The first scheme consists of numerical approximations of the votes, which have been derived from an in-depth analysis of the plate and ball voting processes. The second scheme simplifies the formulation while keeping the same perceptual meaning of the original tensor voting: The stick tensor voting and the stick component of the plate tensor voting must reinforce surfaceness, the plate components of both the plate and ball tensor voting must boost curveness, whereas junctionness must be strengthened by the ball component of the ball tensor voting. Two new parameters have been proposed for the second formulation in order to control the potentially conflictive influence of the stick component of the plate vote and the ball component of the ball vote. Results show that the proposed formulations can be used in applications where efficiency is an issue since they have a complexity of order O(1). Moreover, the second proposed formulation has been shown to be more appropriate than the original tensor voting for estimating saliencies by appropriately setting the two new parameters.
    Full-text · Article · Nov 2011 · IEEE Transactions on Pattern Analysis and Machine Intelligence
  • Source
    • "Even more recent works that improve Furkawa and Ponce's algorithm (e.g. [3]) are not able to produce dense reconstructions . We propose an algorithm to estimate the depth 978-1-61284-162-5/11/$26.00 c 2011 IEEE of the remaining pixels after the initial 3D reconstruction assuming a planar surface model for the textureless surfaces. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Patch cloud based multi-view stereo methods have proven to be an accurate and scalable approach for scene reconstruction. Their applicability, however, is limited due to the semi-dense nature of their reconstruction. We propose a method to generate a dense depth map from a patch cloud by assuming a planar surface model for non-reconstructed areas. We use local evidence to estimate the best fitting plane around missing areas. We then apply a graph cut optimization to select the best plane for each pixel. We demonstrate our approach with a challenging scene containing planar and non-planar surfaces.
    Preview · Conference Paper · Jun 2011
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Tensor voting is a well-known robust technique for extracting perceptual information from clouds of points. This chapter proposes a general methodology to adapt tensor voting to different types of images in the specific context of image structure estimation. This methodology is based on the structural relationships between tensor voting and the so-called structure tensor, which is the most popular technique for image structure estimation. The problematic Gaussian convolution used by the structure tensor is replaced by tensor voting. Afterwards, the results are appropriately rescaled. This methodology is adapted to gray-valued, color, vector- and tensor-valued images. Results show that tensor voting can estimate image structure more appropriately than the structure tensor and also more robustly.
    Full-text · Chapter · Jan 2012
Show more