A 3D Scene Registration Method via Covariance Descriptors and an Evolutionary Stable Strategy Game Theory Solver

Cirujeda, Pol; Cid, Yashin Dicente; Mateo, Xavier; Binefa, Xavier

doi:10.1007/s11263-015-0820-2

Cited by 22 publications

(17 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The complete space of Riesz-wavelet features constitutes a sparse variety, with a lack of characterization for particular texture entities. Following previous research in which we investigated covariance-based descriptors for shape and texture fusion of 3-D surfaces [21], tissue characterization in 3-D CT imaging [22] or 2-D color image categorization [32], we propose to exploit Riesz-wavelet features in their covariance space. The latter can be locally estimated by computing the covariance matrix of the Riesz features in a given ROI.…”

Section: Riesz-covariance 3-d Texture Descriptorsmentioning

confidence: 99%

“…(i) The consideration of voxels within a 3-D region as samples of a multi-dimensional feature distribution implies a loss of structural information, which leads to the robustness to spatial and rotation transformations (due to the steerability of 3-D Riesz-wavelet features) [21], [22]. (ii) The characterization of a set of feature observations by its covariance matrix provides a compact, size and shape independent discriminative signature [32].…”

Section: Riesz-covariance 3-d Texture Descriptorsmentioning

confidence: 99%

“…Overall, the proposed approach provides a flexible framework in which any arbitrary tissue region can be evaluated, regardless of size, shape and connectivity, without the requirement for manual input. The covariance-based descriptors lie on the space of Riemannian geometry, which follows the trend of state-of-the-art work [21], [22], and provides a coherent theoretic framework in order to built classification methods that are sensitive to the topological properties of the described data, e.g. recurrence and locality failure prediction models based on tissue component features.…”

mentioning

confidence: 99%

See 2 more Smart Citations

A 3-D Riesz-Covariance Texture Model for Prediction of Nodule Recurrence in Lung CT

Cirujeda

Cid

Müller

et al. 2016

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

Section: Riesz-covariance 3-d Texture Descriptorsmentioning

confidence: 99%

Section: Riesz-covariance 3-d Texture Descriptorsmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

A 3-D Riesz-Covariance Texture Model for Prediction of Nodule Recurrence in Lung CT

Cirujeda

Cid

Müller

et al. 2016

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

“…Initially, in [1], the Euclidean distances between points are used to measure the consistency between different matches, then the replicator equation is used to estimate their reliabilities on a global scale. In [9], the consistencies of both the descriptors of and Euclidean distances between points are first considered to determine a payoff matrix, the global weights of all the PPMs are then estimated using the infection and immunization dynamics. While invariant features have already been used to represent and match points, it remains a challenge to find other complementary and expressive invariant features for eliminating false matches.…”

Section: Point Match Evaluationmentioning

confidence: 99%

Accurately estimating rigid transformations in registration using a boosting-inspired mechanism

Liu

Martin

et al. 2016

Pattern Recognition

View full text Add to dashboard Cite

Feature extraction and matching provide the basis of many methods for object registration, modeling, retrieval, and recognition. However, this approach typically introduces false matches, due to lack of features, noise, occlusion, and cluttered backgrounds. In registration, these false matches lead to inaccurate estimation of the underlying transformation that brings the overlapping shapes into best possible alignment. In this paper, we propose a novel boosting-inspired method to tackle this challenging task. It includes three key steps: (i) underlying transformation estimation in * Corresponding author. Email: yitian.zhao@bit.edu.cn 1 the weighted least squares sense, (ii) boosting parameter estimation and regularization via Tsallis entropy, and (iii) weight re-estimation and regularization via Shannon entropy and update with a maximum fusion rule. The process is iterated. The final optimal underlying transformation is estimated as a weighted average of the transformations estimated from the latest iterations, with weights given by the boosting parameters. A comparative study based on real shape data shows that the proposed method outperforms four other state-of-the-art methods for evaluating the established point matches, enabling more accurate and stable estimation of the underlying transformation.

show abstract

“…Cirujeda et al [10] presents a descriptor based on the covariance of features, combining shape and color information of 3D surfaces. Multi-scale covariance descriptor (MCOV) has a number of properties including; invariant to spatial rigid transformations, robust to noise and resolution changes and is applicable to 155 characteristic point detection.…”

mentioning

confidence: 99%

Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques

Dupre

Argyriou

Tzimiropoulos

et al. 2016

Information Sciences

View full text Add to dashboard Cite

Please cite this article as: Rob Dupre, Vasileios Argyriou, George Tzimiropoulos, Darrel Greenhill, Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques, Information Sciences (2016Sciences ( ), doi: 10.1016Sciences ( /j.ins.2016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT A C C E P T E D M A N U S C R I P T AbstractIn this paper, the notion of risk analysis within 3D scenes using vision based techniques is introduced. In particular the problem of risk estimation of indoor environments at the scene and object level is considered, with applications in domestic robots and smart homes. To this end, the proposed Risk Estimation Framework is described, which provides a quantified risk score for a given scene.This methodology is extended with the introduction of a novel robust kernel for 3D shape descriptors such as 3D HOG and SIFT3D, which aims to reduce the effects of outliers in the proposed risk recognition methodology. The Physics Behaviour Feature (PBF) is presented, which uses an object's angular velocity obtained using Newtonian physics simulation as a descriptor. Furthermore, an extension of boosting techniques for learning is suggested in the form of the novel Complex and Hyper-Complex Adaboost, which greatly increase the computation efficiency of the original technique. In order to evaluate the proposed robust descriptors an enriched version of the 3D Risk Scenes (3DRS) dataset with extra objects, scenes and meta-data was utilised. A comparative study was conducted demonstrating that the suggested approach outperforms current state-of-the-art descriptors.

show abstract

A 3D Scene Registration Method via Covariance Descriptors and an Evolutionary Stable Strategy Game Theory Solver

Cited by 22 publications

References 27 publications

A 3-D Riesz-Covariance Texture Model for Prediction of Nodule Recurrence in Lung CT

A 3-D Riesz-Covariance Texture Model for Prediction of Nodule Recurrence in Lung CT

Accurately estimating rigid transformations in registration using a boosting-inspired mechanism

Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques

Contact Info

Product

Resources

About