3-D Face Recognition Under Occlusion Using Masked Projection

Alyüz, Neşe; Gökberk, Berk; Akarun, Lale

doi:10.1109/tifs.2013.2256130

Cited by 62 publications

(52 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All in all, the results of our experiments suggest that the FREAK descriptor represents a viable alternative to the SIFT descriptor for the task of 3D face recognition. Even with our simple recognition framework, both descriptors ensured high verification performance; in the case of the UMB-DB even comparable to the state-of-the-art (see, e.g., [23], [24]). When looking at the speed of computation, the FREAK descriptor definitely has an advantage compared to the SIFT descriptor and is also well suited for building recognition systems for low-resource devices such as mobile phones, tablets and alike.…”

Section: Resultsmentioning

confidence: 91%

SIFT vs. FREAK: Assessing the usefulness of two keypoint descriptors for 3D face verification

Krizaj

Dobrišek

Marcetic

et al. 2014

2014 37th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO)

View full text Add to dashboard Cite

Abstract-Many techniques in the area of 3D face recognition rely on local descriptors to characterize the surface-shape information around points of interest (or keypoints) in the 3D images. Despite the fact that a lot of advancements have been made in the area of keypoint descriptors over the last years, the literature on 3D-face recognition for the most part still focuses on established descriptors, such as SIFT and SURF, and largely neglects more recent descriptors, such as the FREAK descriptor. In this paper we try to bridge this gap and assess the usefulness of the FREAK descriptor for the task for 3D face recognition. Of particular interest to us is a direct comparison of the FREAK and SIFT descriptors within a simple verification framework. To evaluate our framework with the two descriptors, we conduct 3D face recognition experiments on the challenging FRGCv2 and UMB-DB databases and show that the FREAK descriptor ensures a very competitive verification performance when compared to the SIFT descriptor, but at a fraction of the computational cost. Our results indicate that the FREAK descriptor is a viable alternative to the SIFT descriptor for the problem of 3D face verification and due to its binary nature is particularly useful for real-timerecognition systems and verification techniques for low-resource devices such as mobile phones, tablets and alike.

show abstract

Section: Resultsmentioning

confidence: 91%

SIFT vs. FREAK: Assessing the usefulness of two keypoint descriptors for 3D face verification

Krizaj

Dobrišek

Marcetic

et al. 2014

2014 37th International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO)

View full text Add to dashboard Cite

show abstract

“…For example, in Passalis et al (2011), facial symmetry is used to handle large pose variations for 3D face recognition in the real world; in Alyüz et al (2013), a masked projection based on subspace analysis techniques is proposed for 3D face recognition under occlusions; in Drira et al (2013), a curve-based shape analysis framework is presented for 3D face recognition under expression changes, occlusions and pose variations. However, all these methods require very sophisticated registration algorithms for automatic pose normalization or occlusion detection and restoration (Alyüz et al 2013;Drira et al 2013).…”

Section: Related Workmentioning

confidence: 99%

“…Although high accuracies have been achieved, very sophisticated registration algorithms are usually indispensable (Kakadiaris et al 2007;Faltemier et al 2008;Al-Osaimi et al 2009;Wang et al 2010;Alyüz et al 2010;Queirolo et al 2010;Spreeuwers 2011;Mohammadzade and Hatzinakos 2013). More recently, 3D face recognition in real biometric applications using scans captured in less controlled or unconstrained conditions has received increasing interests Alyüz et al 2013;Drira et al 2013). In this scenario, 3D face recognition methods are expected to automatically and simultaneously deal with expression changes, occlusions, as well as pose variations.…”

Section: Introductionmentioning

confidence: 99%

Towards 3D Face Recognition in the Real: A Registration-Free Approach Using Fine-Grained Matching of 3D Keypoint Descriptors

et al. 2014

View full text Add to dashboard Cite

Registration algorithms performed on point clouds or range images of face scans have been successfully used for automatic 3D face recognition under expression variations, but have rarely been investigated to solve pose changes and occlusions mainly since that the basic landmarks to initialize coarse alignment are not always available. Recently, local feature-based SIFT-like matchCommunicated by C. Schnörr.ing proves competent to handle all such variations without registration. In this paper, towards 3D face recognition for real-life biometric applications, we significantly extend the SIFT-like matching framework to mesh data and propose a novel approach using fine-grained matching of 3D keypoint descriptors. First, two principal curvature-based 3D keypoint detectors are provided, which can repeatedly identify complementary locations on a face scan where local curvatures are high. Then, a robust 3D local coordinate system is built at each keypoint, which allows extraction of pose-invariant features. Three keypoint descriptors, corresponding to three surface differential quantities, are designed, and their featurelevel fusion is employed to comprehensively describe local shapes of detected keypoints. Finally, we propose a multitask sparse representation based fine-grained matching algorithm, which accounts for the average reconstruction error of probe face descriptors sparsely represented by a large dictionary of gallery descriptors in identification. Our approach is evaluated on the Bosphorus database and achieves rankone recognition rates of 96.56, 98.82, 91.14, and 99.21 % on the entire database, and the expression, pose, and occlusion subsets, respectively. To the best of our knowledge, these are the best results reported so far on this database. Additionally, good generalization ability is also exhibited by the experiments on the FRGC v2.0 database.

show abstract

“…Any part of these images that does not look like part of a face, is considered an occlusion [18] (occluding objects may not touch the face). Normally, in most of the previous works described in [1][2][3][4][5][6], initial detection of occlusions has been very trivial. In the previous works, an initial threshold was formulated on the basis of a trial and error method, just considering the acquisition device.…”

Section: (Iv) Occlusion Detection and Face Restoration:-mentioning

confidence: 99%

“…The authors in [5] have, at first registered the occluded images using an adaptive based registration scheme, restored occlusions using a masked projection scheme and then performed classification using Fisherface projection. Hassen et.…”

Section: Introductionmentioning

confidence: 99%

Robust 3D Face Recognition in Presence of Pose and Partial Occlusions or Missing Parts

Bagchi¹,

Bhattacharjee²,

Nasipuri³

2014

IJFCST

View full text Add to dashboard Cite

In this paper, we propose a robust 3D face recognition system which can handle pose as well as occlusions in real world. The system at first takes as input, a 3D range image, simultaneously registers it using ICP(Iterative Closest Point) algorithm. ICP used in this work, registers facial surfaces to a common model by minimizing distances between a probe model and a gallery model. However the performance of ICP relies heavily on the initial conditions. Hence, it is necessary to provide an initial registration, which will be improved iteratively and finally converge to the best alignment possible. Once the faces are registered, the occlusions are automatically extracted by thresholding the depth map values of the 3D image. After the occluded regions are detected, restoration is done by Principal Component Analysis (PCA). The restored images, after the removal of occlusions, are then fed to the recognition system for classification purpose. Features are extracted from the reconstructed non-occluded face images in the form of face normals. The experimental results which were obtained on the occluded facial images from the Bosphorus 3D face database, illustrate that our occlusion compensation scheme has attained a recognition accuracy of 91.30%.

show abstract

3-D Face Recognition Under Occlusion Using Masked Projection

Cited by 62 publications

References 32 publications

SIFT vs. FREAK: Assessing the usefulness of two keypoint descriptors for 3D face verification

SIFT vs. FREAK: Assessing the usefulness of two keypoint descriptors for 3D face verification

Towards 3D Face Recognition in the Real: A Registration-Free Approach Using Fine-Grained Matching of 3D Keypoint Descriptors

Robust 3D Face Recognition in Presence of Pose and Partial Occlusions or Missing Parts

Contact Info

Product

Resources

About