Real-Time Distance-Dependent Mapping for a Hybrid ToF Multi-Camera Rig

Garcia, Frederic; Aouada, Djamila; Mirbach, Bruno; Ottersten, Björn

doi:10.1109/jstsp.2012.2207090

Cited by 15 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to be a consumer‐accessible depth camera, a major advantages is that the aforementioned depth cameras are a close sensing system that provide real‐time depth and video streams with a well done mapping between the integrated sensors. Therefore, we do not have to deal with the internal transformation, in which distance‐dependent disparity is involve [16], to map the depth data onto the 2D data. In the following, we first motivate the use of a colour guidance image by a quantitative evaluation in which we have considered scenes from the Middlebury stereo dataset [Middlebury Stereo Dataset, http://vision.middlebury.edu/stereo] as well as the two test cases that we have previously generated to illustrate the downside of transforming a colour guidance image to its greyscale version.…”

Section: Resultsmentioning

confidence: 99%

Real‐time depth enhancement by fusion for RGB‐D cameras

Garcia

Aouada

Solignac

et al. 2013

IET Computer Vision

Self Cite

View full text Add to dashboard Cite

Abstract. This paper presents a real-time refinement procedure for depth data acquired by RGB-D cameras. Data from RGB-D cameras suffers from undesired artifacts such as edge inaccuracies or holes due to occlusions or low object remission. In this work, we use recent depth enhancement filters intended for Time-of-Flight cameras, and extend them to structured light based depth cameras, such as the Kinect camera. Thus, given a depth map and its corresponding 2-D image, we correct the depth measurements by separately treating its undesired regions. To that end, we propose specific confidence maps to tackle areas in the scene that require a special treatment. Furthermore, in the case of filtering artifacts, we introduce the use of RGB images as guidance images as an alternative to real-time state-of-the-art fusion filters that use grayscale guidance images. Our experimental results show that the proposed fusion filter provides dense depth maps with corrected erroneous or invalid depth measurements and adjusted depth edges. In addition, we propose a mathematical formulation that enables to use the filter in real-time applications.

show abstract

Section: Resultsmentioning

confidence: 99%

Real‐time depth enhancement by fusion for RGB‐D cameras

Garcia

Aouada

Solignac

et al. 2013

IET Computer Vision

Self Cite

View full text Add to dashboard Cite

show abstract

“…It is based on the assumption that the output has an edge only if the guidance image has one. Garcia et al 14 reduced the weight of unreliable depth samples by analyzing the gradient of LR depth values. Min et al 15 proposed a weighted mode filtering method (WMF) based on joint histogram of the depth candidates, which enforces the result to satisfy the requirement of L1 norm minimization.…”

Section: Filter-based Methodsmentioning

confidence: 99%

Integrated cosparse analysis model with explicit edge inconsistency measurement for guided depth map upsampling

Zuo

et al. 2018

J. Electron. Imag.

View full text Add to dashboard Cite

A low-resolution depth map can be upsampled through the guidance from the registered highresolution color image. This type of method is so-called guided depth map upsampling. Among the existing methods based on Markov random field (MRF), either data-driven or model-based prior is adopted to construct the regularization term. The data-driven prior can implicitly reveal the relation between color-depth image pair by training on external data. The model-based prior provides the anisotropic smoothness constraint guided by highresolution color image. These types of priors can complement each other to solve the ambiguity in guided depth map upsampling. An MRF-based approach is proposed that takes both of them into account to regularize the depth map. Based on analysis sparse coding, the data-driven prior is defined by joint cosparsity on the vectors transformed from color-depth patches using the pair of learned operators. It is based on the assumption that the cosupports of such bimodal image structures computed by the operators are aligned. The edge inconsistency measurement is explicitly calculated, which is embedded into the model-based prior. It can significantly mitigate texture-copying artifacts. The experimental results on Middlebury datasets demonstrate the validity of the proposed method that outperforms seven state-of-the-art approaches.

show abstract

“…Note that this so-called backward warping is far from a trivial task due to the need of determining the distance-dependent disparity between each 2-D and depth pixel correspondences. Herein, we use the real-time backward warping algorithm proposed in [17], an iterative approach that addresses backward warping mapping with an accuracy of half the high-resolution 2-D pixel.…”

Section: Background and Related Workmentioning

confidence: 99%

“…Thereby and in the following depth enhancement techniques by fusion, the low-resolution depth map D has to be upsampled to have the same spatial resolution as I. To this end, nearest-neighbour interpolation might be used [25,48] since alternative data interpolation approaches such as bilinear or bicubic interpolation might introduce non-valid depth measurements among the upsampled edges of the depth map [17]. As in (1), the resulting depth map J JBU is an enhanced version of D that has been upsampled to the 2-D guidance image resolution.…”

Section: Joint Bilateral Upsampling (Jbu)mentioning

confidence: 99%