Motion — Stereo Integration for Depth Estimation

Strecha, Christoph; Gool, Luc J. Van

doi:10.1007/3-540-47967-8_12

Cited by 25 publications

(21 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study several properties affecting the motion integration mechanisms were ignored. For instance, disparity used in binocular experiments is missing, and thus it could be probably possible to evaluate other kinds of models where motion and depth are combined [7,39,30,33]. Another property ignored herein is the contrast.…”

Section: Resultsmentioning

confidence: 99%

Evaluating Motion Estimation Models from Behavioural and Psychophysical Data

Tlapale

Kornprobst

Bouecke

et al. 2012

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 99%

Evaluating Motion Estimation Models from Behavioural and Psychophysical Data

Tlapale

Kornprobst

Bouecke

et al. 2012

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

“…In the earlier works of motion-stereo integration, no matter it is the mere juxtaposition of the results from independent processing of the motion and stereo information (Ayache and Faugeras, 1989;Grosso et al, 1989;Kriegman et al, 1989), where the final estimates of structure were based on some combination of the outputs of these separate processes (coupled loosely together sensus (Clark and Yuille, 1994)), or the tightly coupled approach where the processing of one type of visual information may depend on the presence of another (Balasubramanyam and Snyder, 1991;Li and Duncan, 1993;Shi et al, 1994;Waxman and Duncan, 1993;Zhang and Negahdaripour, 2008), the all but universal assumption is that the overlap in the visual field is used for computing binocular disparity. This assumption remains true in the later approaches with the advent of more sophisticated techniques such as PDE (Strecha and Gool, 2002), variational approach (Huguet and Devernay, 2007;Pons et al, 2007;Williams et al, 2005)and factorization (Ho and Chung, 2000).…”

Section: Literature Reviewmentioning

confidence: 99%

Quasi-Parallax for Nearly Parallel Frontal Eyes

Cheong

Gao

2012

Int J Comput Vis

View full text Add to dashboard Cite

In this paper, we explore how a visual system equipped with a pair of frontally-placed eyes/cameras can rapidly estimate egomotion and depths for the task of locomotion by exploiting the eye topography. We eschew the traditional approach of motion-stereo integration, as finding stereo correspondence is a computationally expensive operation. Instead, we propose a quasi-parallax scheme by pairing appropriate visual rays together, thus obviating the need for stereo correspondence and yet being able to leverage on the redundant information present in the binocular overlap. Our model covers realistic visual systems where the two eyes might deviate from the strictly frontal-parallel configuration, and yet the results show that the advantages of the parallax-based approach are retained. In particular, it offers better disambiguation of translation and rotation over conventional two-frame structure from motion approaches, despite not having views covering diametrically opposing directions like that of spherical eyes or laterally-placed eyes. The rapid processing that such scheme entails seems to offer a more realizable and useful alternative for depth recovery during locomotion.

show abstract

“…Others have focused on defining appropriate temporal integration windows, e.g., as part of correspondence [28] or simply reinforce disparity estimates from the previous frame using optical flow [9]. Further, combined stereo and motion estimation has been formulated in terms of PDEs [25,11] as well as MRFs [27,15]. Still other work has used direct methods for integrated recovery of structure and egomotion [24,17].…”

Section: Introductionmentioning

confidence: 99%