Exploring Facial Expression Effects in 3D Face Recognition Using Partial ICP

Size: px
Start display at page:

Download "Exploring Facial Expression Effects in 3D Face Recognition Using Partial ICP"

Transcription

1 Exploring Facial Expression Effects in 3D Face Recognition Using Partial ICP Yueming Wang 1, Gang Pan 1,, Zhaohui Wu 1, and Yigang Wang 2 1 Dept. of Computer Science, Zhejiang University, Hangzhou, , China Tel: Virtual Reality Lab, Hangzhou Dianzi University, Hangzhou, , China {ymingwang, gpan}@zju.edu.cn Abstract. This paper investigates facial expression effects in face recognition from 3D shape using partial ICP. The partial ICP method could implicitly and dynamically extract the rigid parts of facial surface by selecting a part of nearest points pairs to calculate dissimilarity measure during registration of facial surfaces. The method is expected to be able to get much better performance than other methods in 3D face recognition under expression variation for its dynamic extraction of rigid parts of facial surface at the same time of matching. We also present an effective method for coarse alignment of facial shape, which is fully automatic. Experiments on 3D face database of 360 models with 40 subjects, 9 scans with four different kinds of expression for each subject, show partial ICP is very promising, compared with PCA baseline. 1 Introduction Automatic face recognition has been studied extensively over the past decade. Most efforts have been made for face recognition from 2D images[1] and a few approaches exploited 3D information [2, 3, 5, 6, 7, 8, 9, 10]. Although the 2D face recognition system has good performance under constrained conditions, since the 2D image essentially is a projection of the 3D human face, it is still challenged by changes in illumination, pose and expression [1, 17]. Utilizing 3D information can improve the system performance[17,7] due to its explicit representation of facial surface. However, facial expression is still a big challenge even using 3D data in face recognition because in fact facial surface is a non-rigid object. Some efforts have been made to conquer the problem. C.S.Chua et.al [5] extracted rigid parts of facial surface by a Gaussian model after registering the face range data with varying expression. These rigid parts were used to create a model library for indexing. After enrolling four scans for each subject, voting based on point signature was employed for recognition. It was reported that 100% recognition rate was obtained for total six persons. However, the model database is too small.. Furthermore, by Gaussian Distribution, The authors are grateful for the grants from the National Science Foundation of China ( , ) and Program for New Century Excellent Talents in University (NCET ). Corresponding author. P.J. Narayanan et al. (Eds.): ACCV 2006, LNCS 3851, pp , c Springer-Verlag Berlin Heidelberg 2006

2 582 Y. Wang et al. almost all extracted rigid parts of models only discarded mouth from full facial surface. Some expression may deform other areas of facial surface such as cheek. K.Chang et.al [11] proposed a local region approach to coping with expression variation in 3D face recognition. The algorithm is based on traditional ICP after finding nose area of facial surface. On a database with about 355 subjects and D models with seven different expressions, an average rank-1 rate 77.1% was obtained. The algorithm improved the recognition performance, compared with ICP-baseline method using complete facial shape. Their work treated nose area as the rigid region under varying expressions. But under certain expressions, parts of the nose still show some deformations. A.M.Bronstein et.al [12] reported a 3D face recognition approach based on a representation of the facial surface that was invariant to isometric deformations resulting from expression variation. However, geodesic distance is definitely variant when facial surface with a mouth open expression. There is a common assumption that though the face shape of the same person may change, sometimes greatly, due to various facial expressions, still there are regions which will keep their shape and position or be subjected to much less deformation among different expressions. If these regions can be identified, the 3D non-rigid face recognition problem can be reduced to the rigid case[5, 11]. However, there may be no large uniform subset of the face that is perfectly shape invariant across a broad range of normal expressions, and the deformation of facial surface of a certain person may be not as same as others with the similar expressions at all time. Figure 1(a) shows some deformation images of facial surface with three different expressions, smile, surprise and sad. The deformation image is obtained as follows. Registering neutral expression face with non-neutral face of same subject by nose area and subtracting the former from the latter along the depth value, we call the difference map deformation image, which indicates the deformation extent of surface region with certain expression relative to neutral facial surface. And the darker in the figure indicates more deformation and the lighter means less deformation. From the deformation images, it can be seen that: (1) For a subject, almost no fixed large parts of facial surface are invariant along three expressions. Shown in left four columns in Fig.1(a), smile expression leads to shape deformation of mouth and cheek, surprise affects mouth and sad even changes the shape of nose and forehead area slightly. (2) Comparing two subjects in Fig.1(a), the same expression of different subject affects different regions. Sad in left person changes shape of forehead, while affects mainly eyebrow in right person. Thus, just extracting and matching the same relatively rigid parts for all facial surfaces is only a choice to solve the expression problem and may not be perfect. In this paper, after analysis of iterative closest point(icp), we give the partial ICP for 3D facial shape recognition which can implicitly extract variant rigid regions of the face according to deformation extent under different expression during matching. The method does a proper selection of nearest points pairs

3 Exploring Facial Expression Effects in 3D Face Recognition 583 (a) (b) Fig. 1. (a) Deformation images for two subjects. The darker indicates the more deformation. (b) Discarded area in facial surface with different p-rate=0.9,0.7,0.2 (5th,6th,7th columns respectively). Regions in red indicate the removed parts. Removed areas are not fixed between facial surfaces under different expressions. to calculate RMS when using ICP to match two surfaces. When applied the method to three expressions, smile, surprise and sad in our experiments, 96.88% rank-one matching rate is obtained. We also implement the PCA-based 3D face recognition as a baseline algorithm. This paper is organized as follows: Sec. 2 analyzes the ICP algorithm and presents our method of implicitly extracting rigid parts of facial surface. Sec. 3 describes the data preparing. The experimental results and conclusions are in Sec. 4 and Sec. 5 respectively. 2 Analysis of Iterative Closest Point (ICP) The ICP algorithm, developed by Besl and Mckay [16], is used to register the point sets by an iterative procedure which is widely used in field of 3D rigid

4 584 Y. Wang et al. object registration. Let point set P 1 = {p 1 1,,p1 M } and point set P 2 = {p 2 1,,p 2 N }. The ICP algorithm is summarized as: 1. P 2 (0) = P 2, l=0 2. Do 3. For each point p 2 i in P 2 (l) 4. Find the closest point y i in P 1 5. End For 6. The closest points y i form a new point set Y(l) where 7. the pairs of points {(p 2 1,y 1 ),,(p 2 N,y N )} 8. describe the correspondences between P 1 and P 2 (l). 9. If registration error E between P 1 and 10. P 2 (l) istoolarge 11. Compute transformation T(l) between (P 2 (l), Y(l)), 12. including translation and rotation. 13. Apply transformation P 2 (l +1)=T (l) P 2 (l),l=l Else 15. Stop 16. End If 17. While P 2 (l +1) P 2 (l) >threshold where point y k in set Y(l) denotes the closest point in P 1 to the point p 2 k(l) in P 2 (l) and the registration error between P 1 and P 2 (l) is E = 1 N N y k p 2 k(l) 2 (1) k For convergence of ICP, a coarse registration step usually is carried out before the iterative process. Generally, in ICP-based 3D face recognition, two facial surfaces are registered by the above method, then the value of E computed in the last time of iterative steps is treated as dissimilarity measure of two faces. When matching two facial surfaces with different expressions, the difference between the pairs of nearest points may become large due to shape deformation which may have a large effect when performing least-squares minimization and E is no longer accurate as a dissimilarity metric. Thus, there is a significant performance drop by ICP-Based method in 3D face recognition when expression varies between gallery and probe, from average 93.6% to 61.1%, as reported by K.Chang [11]. If only those pairs of points with relatively less deformation can be selected as input of calculation of E, the registration error E may be still able to distinguish different subjects while remain small when matching models of same subjects with different expression. While the traditional ICP-based method uses all point pairs in computing transformation T (l) ande [11], we do it by selecting parts of the point pairs. After sorting the distances of pairs of points in increasing order, we reject the worst n% of pairs based on distance in each pair. That is, only first (1-n%) part of distances and corresponding point pairs from sorted distances are chosen

5 Exploring Facial Expression Effects in 3D Face Recognition 585 to compute transformation E and T (l). Considering the last E that is used as dissimilarity measure of matching, discarding n% of pairs means removing those points in non-rigid region of facial surface. Thus, it is a implicit method to extract points in rigid parts of facial surface to register and match and the rigid parts extracted are varied according to deformation of facial surface among different matching models. We denote it partial ICP for 3D face recognition approach and call (1-n%) p-rate. Figue 1(b) shows some deformation images in which the areas in red indicate those removed by setting different p-rate. From the removed area, it could be seen that red regions completely come from darker area in deformation images. When p-rate equals 0.7, 70 percent of face area is kept to match and most nonrigid parts are discarded. Thus, the method is expected to be able to get much better performance in 3D face recognition with expressions than other rigidparts-based methods for its dynamically extracting rigid areas of facial surface at the same time of matching. 3 Data Preparing Considering the convergence problem of partial ICP, we firstly transform all models into a canonical coordinate system by finding the symmetric plane of facial surface and detecting two fiducial points, nose tip and nose base. Then, facial regions for all models are well extracted by trimming face mesh models with a elliptical cylinder which coarsely extracts same facial regions for all models, as shown in Fig.1. After trimming face models, following two strategies are applied: (1) To compensate for the effect of resolution,we simplify trimmed models using mesh optimization [15]. Then, all facial surface meshes put into experiments have about 2000 vertices. (2) When finding nearest point pairs between two point face meshes in partial ICP, the nearest distance from point to surface is computed instead of nearest distance between vertices of meshes. The details of alignment and trimming are described as follows. 3.1 Transforming to the Canonical Coordinate System Suppose central profile passes through nose tip, nose base and is in the symmetric plane of the facial surface. From profile, we identify following information: nose tip p nt, nose base p nb and direction of symmetric plane d s. Obviously, six degrees of freedom of facial surface can be fixed by p nt, p nb and d s. After detecting these information for each facial surface, all models can be coarsely registered in a canonical coordinate frame. Finding Facial Central Profile. We apply our early work [13, 14] to detect the curve of the central profile of facial surface, as reviewed briefly as follows. Let S(p i ) denotes a point set of facial surface, where p i is a point in the set and S m (p m i ) denotes its mirror to some certain plane, where pm i is corresponding

6 586 Y. Wang et al. mirrored point of p i.whens m (p m i ) has been registered to S(p i), S m (p m i )is transformed into S m (p m i ). From Sm (p m i ) and S(p i), we can fit symmetric plane of facial surface from point set A(x i ), where each point x i obtained by: x i =(p i + p m i)/2 (2) We use the basic idea of the ICP to get a registration between facial surface and its mirror and find the symmetric plane by equation 2. Finally, we calculate the intersection of symmetric plane and the triangulated surface of S(p i )toget the central profile, shown in Fig.2. y x z (a) (b) (c) (d) (e) (f) Fig. 2. Symmetric plane detection and profile finding using ICP. (a) original model, (b) mirrored model, (c) detected symmetric plane, (e) profile and symmetric plane seen from another view, (f) determination of nose tip and nose base in profile. Locating Nose Tip and Nose Base. Since central profile curve passes through nose tip and nose base, we locate their positions based on the curve. Let C denotes the central profile curve extracted, l e denotes the line through both end points of the curve C, p nt and p nb denote nose tip point and nose base point respectively. Before location, we suppose following assumption hold up. (1) The nose tip p nt is a point on the curve C, with the maximum distance to the line l e. (2) The nose base p nb is a point on the profile curve C, and is the first distance extremum point to the line l e from p nt to forehead, as shown in Fig. 2(f). It can be formalized as: p nt = argmax p C dist(p, l e ) (3) L = {p p C, y p >y pnt,dist (p, l e )=0} (4) p nb = argmin p L (y p ) (5) where dist(, ) is the Euclidean distance function from a point to a line segment, y p is y-axis coordinate of point p, dist (, ) denotes first derivative of Euclidean distance to the point position at line l e extend to forehead. If facial surface is sampled only frontal view discarding hair and neck area, our assumption is appropriate so that p nb and p nt can be located accurately.

7 Exploring Facial Expression Effects in 3D Face Recognition 587 However, some certain samples of facial surface may be grotesque in shape which doesn t keep the assumption. To date we have never encountered a model on which failure happen in our experimental data set. Aligning Model. Given nose tip p nt, nose base p nb and normal direction v sp of symmetric plane for each model, a canonical coordinate system of all models can be determined. Subtracting p nb from p nt, we get unit vector v y after normalized. Taking p nt as the origin, v sp as x-axis, v y as y-axis, a new right-hand coordinate system is defined. Then all models are registered by transforming facial surface into the new coordinate system. Furthermore, we rotate the model 20 degree around x-axis counterclockwise in the new coordinate system for non-duplicate happened in projecting depth to x-y plane used in PCA-based face recognition. Some results are shown in Fig.3 (a). All our experiments are based on the aligned models. (a) (b) Fig. 3. (a) The canonical coordinate system for aligned models. Axis in red is z-axis, blue is y-axis, green is x-axis. (b) Face models acquired by InSpeck 3D MEGA Capturor DF. Each row shows 9 scans of one subject. The models in first row are rendering with texture. 3.2 Trimming Models Given aligned model mesh, facial regions can be extracted by removing those points and triangles of facial surfaces in the outer of following elliptical cylinder: (x x 1 ) 2 a 2 + (y y 1) 2 b 2 =1 (6) Since all models are in a canonical coordinate, the facial regions of all models produced by above equation are not only full frontal area, but also roughly same between models which is an important condition for our partial ICP method. For consistency, we set parameters as x 1 =0, y 1 =20, a=60 and b=80 which works well for all 360 models in our experiments. 4 Experiments 4.1 Data Acquisition Experiments use the 3D facial expression database ZJU-3DFED, collected by the authors. In ZJU-3DFED database, there are 40 different subjects, nine scans

8 588 Y. Wang et al. for each, and total 360 scans. Each subject has two scans with smile expression, two scans with surprise expression, 2 scans with sad expression and 3 scans with neutral expression. Facial surfaces of same subject with same expression are slightly different in extent. All face models are acquired by InSpeck 3D MEGA Capturor DF[18]. The facial models are in triangular mesh. We manually cut out the parts out of the face regions from the original model data and this is the only manual work in our whole works. Each facial mesh then have about points and triangles and the resolution is 0.04 mm. After mesh simplification [15], each scan has about 2000 points and about 4000 triangles. Figure 3(b) shows 3 subjects and 27 scans of face models. The face models have both shape and texture information, we only use shape information in the experiments. We put one neutral expression face model for each subject into gallery and the other 320 scans are classified into 4 probe sets. All 80 smile scans form smile-set, so do surprise scans, sad scans and neutral scans. A special probe set composed of the 320 scans is made for whole recognition results, called whole-set. 4.2 Results by Partial ICP with Different p-rate Twelve different values of p-rate { 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,0.6, 0.7, 0.8, 0.9, 0.95, 1} are tested in our experiments. Additionally, we also consider a extreme instance that only a pair of nearest points is input into calculation of dissimilarity measure E after ICP process which use all points pair in iterative. The results are in Fig. 4. Rank-1 rate Smile Surprise Sad Neutral Whole p-rate (a) Rank-1 Rate Smile Surprise 0.85 Sad Neutral Whole p-rate (b) Fig. 4. (a) Rank-1 recognition rate of different p-rate with partial ICP method on five probe sets. (b) zooming out part of (a). From Fig.4, it can be seen that: (1) For three non-neutral probe sets, none of them has a rank-1 rate larger than 90% when p-rate equals 1(that is same as the traditional ICP-based matching method). But when setting p-rate value between 0.1 and 0.95, none of them has a rank-1 rate smaller than 90%. The largest improvements of rank-1 rate of three non-neutral sets are 7.5% for smile-set, 11.25% for surprise-set, 10% for sad set.

9 Exploring Facial Expression Effects in 3D Face Recognition 589 (2) Both highest rank-1 rates of smile-set and sad-set are 96.25% and obtained at p-rate =0.8 while that of surprise-set are 97.5% and obtained at p-rate= 0.1 or p-rate= 0.2. It is partly due to facial surface with surprise expression has a larger deformation area than the other expressions. (3) When setting p-rate as 0.1 which means 90% of facial surface is removed before matching using partial ICP method, an average rank-1 rate 94.17% is still reached on non-neutral probe sets. It is a cue indicating that small parts of facial surface still have enough information for recognition if nice extraction is performed. (4) As a whole, our method get an average rank-1 rate 95% on three nonneutral probe sets when p-rate=0.2 and 96.88% on whole-set. 4.3 PCA v.s. Partial ICP PCA-based method is implemented in our experiments for comparison. After models are trimmed, PCA-based method can easy be applied to 3D face recognition by projecting the trimmed models to x-y plane. We use the first 40 eigenvectors when test PCA-based method which hold 96.46% energy. We compare the performance between PCA-based method and partial ICP method on all five probe sets. The results are shown in Fig.5. Fig. 5. Rank-1 rate: PCA v.s. partial ICP The partial ICP method outperforms PCA-based-method on rank-1 performance among all probe sets. PCA-based-method is well known that is sensitive to noise. On all non-neutral expression probe set, PCA-based-method get average rank-1 rate 75.41% and the worst rank-1 rate 65% on surprise-set because shape deformation with different expressions act as a role of noise in a way. Between neutral and non-neutral expression probe sets, the rank-1 rate drop from 92.5% to an average 75.41% with PCA-Base method in recognition. The partial ICP with p-rate=1 gets a whole rank-1 recognition rate 89.69%. The partial ICP with best p-rate obtains an average rank-1 rate 96.88% on all probe sets. By well selecting the p-rate, partial ICP method is insensitive to expression variant in 3D face recognition.

10 590 Y. Wang et al. 5 Conclusion We propose a method, partial ICP method, which is capable of dynamically extracting rigid parts of facial surface. The extraction is completely dependent on the deformation extent of facial surface and extracted areas are varied between different expressions. Based on partial ICP, we perform several experiments for 3D face recognition on a database with 360 models. A rank-1 rate 96.88% demonstrate the high performance of our method in 3D face recognition with different expressions. The experimental results also show that our method significantly outperforms PCA-based method. References 1. W.Zhao, R.Chellappa, P.J.Phillips, A.Rosenfeld. Face recognition: a literature survey. ACM Computing Surveys, 35(4): , J.C.Lee, E.Milios. Matching range images of human faces. Proc. IEEE ICCV, p , G.G.Gordon. Face recognition from depth maps and surface curvature. SPIE Conf. on Geometric Methods in Computer Vision, 1570: , C.S.Chua, R.Jarvis. Point signatures: A new representation for 3D object recognition. IJCV, 25(1):63-85, C.S.Chua, F.Han, Y.K.Ho. 3D Human Face Recognition Using Point Signature. IEEE FG 00, pp , M.W.Lee, S.Ranganath. Pose-invariant face recognition using a 3D deformable model. Pattern Recognition, 36(8): , V.Blanz, S.Romdhani, T.Vetter. Face identification across different poses and illumination with a 3D morphable model. Int l Conf. on FG, p , C.Beumier, M.Acheroy. Automatic 3D face authentication. Image Vision Computing, 18(4): , W. Zhao, R. Chellappa. Illumination-insensitive face recognition using symmetric shape-form-shading. Proc. IEEE ICCV, 1: , G. Pan, Z. Wu, and Y. Pan, Automatic 3D face verification from range data, in Proc. IEEE ICASSP, vol.3, pp , K.Chang, K.Bowyer, P.Flynn. Effects on Facial Expression in 3D Face Recognition, Proc.oftheSPIE, Volume 5779, pp , A.M.Bronstein, M.M.Bronstein, R.Kimmel. Expression-invariant 3D face recognition. Proc. AVBPA 03, LNCS, vol.2688, 62-70, Yijun Wu, Gang Pan, Zhaohui Wu. Face Authentication based on Multiple Profiles Extracted from Range Data. Proc. AVBPA 03, LNCS, vol.2688, pp , Gang Pan, Zhaohui Wu, 3D Face Recognition from Range Data, Int l Journal of Image and Graphics, 5(3): , H.Hoppe, T.DeRose, T.Duchamp, J.McDonald and W. Stuetzle, Mesh optimization, Computer Graphics(SIGGRAPH 93 Proceedings), 27:19-26, Auguest, P.J.Besl, N.D.McKay, A method for registration of 3-D shapes, IEEE Trans.Pattern Anal.Mach.Intell. 14: , Face Recognition Vendor Test 2002, InSpeck Inc.,

Face Authentication Based on Multiple Profiles Extracted from Range Data

Face Authentication Based on Multiple Profiles Extracted from Range Data Face Authentication Based on Multiple Profiles Extracted from Range Data Yijun Wu, Gang Pan, and Zhaohui Wu Department of Computer Science and Engineering Zhejiang University, Hangzhou, 310027, P. R. China

More information

Face Alignment Under Various Poses and Expressions

Face Alignment Under Various Poses and Expressions Face Alignment Under Various Poses and Expressions Shengjun Xin and Haizhou Ai Computer Science and Technology Department, Tsinghua University, Beijing 100084, China ahz@mail.tsinghua.edu.cn Abstract.

More information

Algorithm research of 3D point cloud registration based on iterative closest point 1

Algorithm research of 3D point cloud registration based on iterative closest point 1 Acta Technica 62, No. 3B/2017, 189 196 c 2017 Institute of Thermomechanics CAS, v.v.i. Algorithm research of 3D point cloud registration based on iterative closest point 1 Qian Gao 2, Yujian Wang 2,3,

More information

The Novel Approach for 3D Face Recognition Using Simple Preprocessing Method

The Novel Approach for 3D Face Recognition Using Simple Preprocessing Method The Novel Approach for 3D Face Recognition Using Simple Preprocessing Method Parvin Aminnejad 1, Ahmad Ayatollahi 2, Siamak Aminnejad 3, Reihaneh Asghari Abstract In this work, we presented a novel approach

More information

New Experiments on ICP-Based 3D Face Recognition and Authentication

New Experiments on ICP-Based 3D Face Recognition and Authentication New Experiments on ICP-Based 3D Face Recognition and Authentication Boulbaba Ben Amor Boulbaba.Ben-Amor@ec-lyon.fr Liming Chen Liming.Chen@ec-lyon.fr Mohsen Ardabilian Mohsen.Ardabilian@ec-lyon.fr Abstract

More information

Face Recognition At-a-Distance Based on Sparse-Stereo Reconstruction

Face Recognition At-a-Distance Based on Sparse-Stereo Reconstruction Face Recognition At-a-Distance Based on Sparse-Stereo Reconstruction Ham Rara, Shireen Elhabian, Asem Ali University of Louisville Louisville, KY {hmrara01,syelha01,amali003}@louisville.edu Mike Miller,

More information

Structured Light II. Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov

Structured Light II. Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov Structured Light II Johannes Köhler Johannes.koehler@dfki.de Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov Introduction Previous lecture: Structured Light I Active Scanning Camera/emitter

More information

State of The Art In 3D Face Recognition

State of The Art In 3D Face Recognition State of The Art In 3D Face Recognition Index 1 FROM 2D TO 3D 3 2 SHORT BACKGROUND 4 2.1 THE MOST INTERESTING 3D RECOGNITION SYSTEMS 4 2.1.1 FACE RECOGNITION USING RANGE IMAGES [1] 4 2.1.2 FACE RECOGNITION

More information

Enhancing 3D Face Recognition By Mimics Segmentation

Enhancing 3D Face Recognition By Mimics Segmentation Enhancing 3D Face Recognition By Mimics Segmentation Boulbaba Ben Amor, Mohsen Ardabilian, and Liming Chen MI Department, LIRIS Laboratory, CNRS 5205 Ecole Centrale de Lyon, 36 av. Guy de Collongue, 69134

More information

Component-based Registration with Curvature Descriptors for Expression Insensitive 3D Face Recognition

Component-based Registration with Curvature Descriptors for Expression Insensitive 3D Face Recognition Component-based Registration with Curvature Descriptors for Expression Insensitive 3D Face Recognition Neşe Alyüz Boğaziçi University Computer Engineering Dept. Istanbul, Turkey nese.alyuz@boun.edu.tr

More information

Integrating Range and Texture Information for 3D Face Recognition

Integrating Range and Texture Information for 3D Face Recognition Integrating Range and Texture Information for 3D Face Recognition Xiaoguang Lu and Anil K. Jain Dept. of Computer Science & Engineering Michigan State University East Lansing, MI 48824 {Lvxiaogu, jain}@cse.msu.edu

More information

Three-Dimensional Face Recognition: A Fishersurface Approach

Three-Dimensional Face Recognition: A Fishersurface Approach Three-Dimensional Face Recognition: A Fishersurface Approach Thomas Heseltine, Nick Pears, Jim Austin Department of Computer Science, The University of York, United Kingdom Abstract. Previous work has

More information

3D Computer Vision. Structured Light II. Prof. Didier Stricker. Kaiserlautern University.

3D Computer Vision. Structured Light II. Prof. Didier Stricker. Kaiserlautern University. 3D Computer Vision Structured Light II Prof. Didier Stricker Kaiserlautern University http://ags.cs.uni-kl.de/ DFKI Deutsches Forschungszentrum für Künstliche Intelligenz http://av.dfki.de 1 Introduction

More information

AAM Based Facial Feature Tracking with Kinect

AAM Based Facial Feature Tracking with Kinect BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 15, No 3 Sofia 2015 Print ISSN: 1311-9702; Online ISSN: 1314-4081 DOI: 10.1515/cait-2015-0046 AAM Based Facial Feature Tracking

More information

Landmark Detection on 3D Face Scans by Facial Model Registration

Landmark Detection on 3D Face Scans by Facial Model Registration Landmark Detection on 3D Face Scans by Facial Model Registration Tristan Whitmarsh 1, Remco C. Veltkamp 2, Michela Spagnuolo 1 Simone Marini 1, Frank ter Haar 2 1 IMATI-CNR, Genoa, Italy 2 Dept. Computer

More information

Chapter 3 Image Registration. Chapter 3 Image Registration

Chapter 3 Image Registration. Chapter 3 Image Registration Chapter 3 Image Registration Distributed Algorithms for Introduction (1) Definition: Image Registration Input: 2 images of the same scene but taken from different perspectives Goal: Identify transformation

More information

Pose Normalization for Robust Face Recognition Based on Statistical Affine Transformation

Pose Normalization for Robust Face Recognition Based on Statistical Affine Transformation Pose Normalization for Robust Face Recognition Based on Statistical Affine Transformation Xiujuan Chai 1, 2, Shiguang Shan 2, Wen Gao 1, 2 1 Vilab, Computer College, Harbin Institute of Technology, Harbin,

More information

Structured Light II. Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov

Structured Light II. Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov Structured Light II Johannes Köhler Johannes.koehler@dfki.de Thanks to Ronen Gvili, Szymon Rusinkiewicz and Maks Ovsjanikov Introduction Previous lecture: Structured Light I Active Scanning Camera/emitter

More information

Component-based Face Recognition with 3D Morphable Models

Component-based Face Recognition with 3D Morphable Models Component-based Face Recognition with 3D Morphable Models B. Weyrauch J. Huang benjamin.weyrauch@vitronic.com jenniferhuang@alum.mit.edu Center for Biological and Center for Biological and Computational

More information

I CP Fusion Techniques for 3D Face Recognition

I CP Fusion Techniques for 3D Face Recognition I CP Fusion Techniques for 3D Face Recognition Robert McKeon University of Notre Dame Notre Dame, Indiana, USA rmckeon@nd.edu Patrick Flynn University of Notre Dame Notre Dame, Indiana, USA flynn@nd.edu

More information

Figure 1. Example sample for fabric mask. In the second column, the mask is worn on the face. The picture is taken from [5].

Figure 1. Example sample for fabric mask. In the second column, the mask is worn on the face. The picture is taken from [5]. ON THE VULNERABILITY OF FACE RECOGNITION SYSTEMS TO SPOOFING MASK ATTACKS Neslihan Kose, Jean-Luc Dugelay Multimedia Department, EURECOM, Sophia-Antipolis, France {neslihan.kose, jean-luc.dugelay}@eurecom.fr

More information

Selection of Location, Frequency and Orientation Parameters of 2D Gabor Wavelets for Face Recognition

Selection of Location, Frequency and Orientation Parameters of 2D Gabor Wavelets for Face Recognition Selection of Location, Frequency and Orientation Parameters of 2D Gabor Wavelets for Face Recognition Berk Gökberk, M.O. İrfanoğlu, Lale Akarun, and Ethem Alpaydın Boğaziçi University, Department of Computer

More information

Nonrigid Surface Modelling. and Fast Recovery. Department of Computer Science and Engineering. Committee: Prof. Leo J. Jia and Prof. K. H.

Nonrigid Surface Modelling. and Fast Recovery. Department of Computer Science and Engineering. Committee: Prof. Leo J. Jia and Prof. K. H. Nonrigid Surface Modelling and Fast Recovery Zhu Jianke Supervisor: Prof. Michael R. Lyu Committee: Prof. Leo J. Jia and Prof. K. H. Wong Department of Computer Science and Engineering May 11, 2007 1 2

More information

Surface Registration. Gianpaolo Palma

Surface Registration. Gianpaolo Palma Surface Registration Gianpaolo Palma The problem 3D scanning generates multiple range images Each contain 3D points for different parts of the model in the local coordinates of the scanner Find a rigid

More information

On Modeling Variations for Face Authentication

On Modeling Variations for Face Authentication On Modeling Variations for Face Authentication Xiaoming Liu Tsuhan Chen B.V.K. Vijaya Kumar Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA 15213 xiaoming@andrew.cmu.edu

More information

Face Tracking. Synonyms. Definition. Main Body Text. Amit K. Roy-Chowdhury and Yilei Xu. Facial Motion Estimation

Face Tracking. Synonyms. Definition. Main Body Text. Amit K. Roy-Chowdhury and Yilei Xu. Facial Motion Estimation Face Tracking Amit K. Roy-Chowdhury and Yilei Xu Department of Electrical Engineering, University of California, Riverside, CA 92521, USA {amitrc,yxu}@ee.ucr.edu Synonyms Facial Motion Estimation Definition

More information

A 3D Point Cloud Registration Algorithm based on Feature Points

A 3D Point Cloud Registration Algorithm based on Feature Points International Conference on Information Sciences, Machinery, Materials and Energy (ICISMME 2015) A 3D Point Cloud Registration Algorithm based on Feature Points Yi Ren 1, 2, a, Fucai Zhou 1, b 1 School

More information

3D Face Identification - Experiments Towards a Large Gallery

3D Face Identification - Experiments Towards a Large Gallery 3D Face Identification - Experiments Towards a Large Gallery Dirk Colbry a, Folarin Oki b, George Stockman b a Arizona State University, School of Computing and Informatics, Tempe, AZ 85287-8809 USA b

More information

Low Dimensional Surface Parameterisation with Applications in Biometrics

Low Dimensional Surface Parameterisation with Applications in Biometrics Low Dimensional Surface Parameterisation with Applications in Biometrics Wei Quan, Bogdan J. Matuszewski, Lik-Kwan Shark, Djamel Ait-Boudaoud Applied Digital Signal and Image Processing Centre University

More information

DETC D FACE RECOGNITION UNDER ISOMETRIC EXPRESSION DEFORMATIONS

DETC D FACE RECOGNITION UNDER ISOMETRIC EXPRESSION DEFORMATIONS Proceedings of the ASME 2014 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference IDETC/CIE 2014 August 17-20, 2014, Buffalo, New York, USA DETC2014-34449

More information

Range Image Registration with Edge Detection in Spherical Coordinates

Range Image Registration with Edge Detection in Spherical Coordinates Range Image Registration with Edge Detection in Spherical Coordinates Olcay Sertel 1 and Cem Ünsalan2 Computer Vision Research Laboratory 1 Department of Computer Engineering 2 Department of Electrical

More information

Model-based segmentation and recognition from range data

Model-based segmentation and recognition from range data Model-based segmentation and recognition from range data Jan Boehm Institute for Photogrammetry Universität Stuttgart Germany Keywords: range image, segmentation, object recognition, CAD ABSTRACT This

More information

Automatic 3D Face Recognition Combining Global Geometric Features with Local Shape Variation Information

Automatic 3D Face Recognition Combining Global Geometric Features with Local Shape Variation Information Automatic 3D Face Recognition Combining Global Geometric Features with Local Shape Variation Information Chenghua Xu 1, Yunhong Wang 1, Tieniu Tan 1, Long Quan 2 1 Center for Biometric Authentication and

More information

EXPLOITING 3D FACES IN BIOMETRIC FORENSIC RECOGNITION

EXPLOITING 3D FACES IN BIOMETRIC FORENSIC RECOGNITION 18th European Signal Processing Conference (EUSIPCO-2010) Aalborg, Denmark, August 23-27, 2010 EXPLOITING 3D FACES IN BIOMETRIC FORENSIC RECOGNITION Marinella Cadoni, Andrea Lagorio, Enrico Grosso Massimo

More information

3D Face Recognition. Anil K. Jain. Dept. of Computer Science & Engineering Michigan State University.

3D Face Recognition. Anil K. Jain. Dept. of Computer Science & Engineering Michigan State University. 3D Face Recognition Anil K. Jain Dept. of Computer Science & Engineering Michigan State University http://biometrics.cse.msu.edu Face Recognition 1959? 1960 1972 1973 Face detection using OpenCV Viola-Jones

More information

Detecting Multiple Symmetries with Extended SIFT

Detecting Multiple Symmetries with Extended SIFT 1 Detecting Multiple Symmetries with Extended SIFT 2 3 Anonymous ACCV submission Paper ID 388 4 5 6 7 8 9 10 11 12 13 14 15 16 Abstract. This paper describes an effective method for detecting multiple

More information

Illumination invariant face recognition and impostor rejection using different MINACE filter algorithms

Illumination invariant face recognition and impostor rejection using different MINACE filter algorithms Illumination invariant face recognition and impostor rejection using different MINACE filter algorithms Rohit Patnaik and David Casasent Dept. of Electrical and Computer Engineering, Carnegie Mellon University,

More information

Enhanced Active Shape Models with Global Texture Constraints for Image Analysis

Enhanced Active Shape Models with Global Texture Constraints for Image Analysis Enhanced Active Shape Models with Global Texture Constraints for Image Analysis Shiguang Shan, Wen Gao, Wei Wang, Debin Zhao, Baocai Yin Institute of Computing Technology, Chinese Academy of Sciences,

More information

EI3D: Expression-Invariant 3D Face Recognition based on Feature and Shape Matching

EI3D: Expression-Invariant 3D Face Recognition based on Feature and Shape Matching 1 Pattern Recognition Letters journal homepage: www.elsevier.com EI3D: Expression-Invariant 3D Face Recognition based on Feature and Shape Matching Yulan Guo a,, Yinjie Lei b, Li Liu c, Yan Wang d, Mohammed

More information

3D Face Recognition using Mapped Depth Images

3D Face Recognition using Mapped Depth Images 3D Face Recognition using Mapped Depth Images Gang Pan, Shi Han, Zhaohui Wu and Yueming Wang Department of Computer Science Zhejiang University, Hangzhou, 310027, P.R.China {gpan, hanshi, wzh, ymingwang}@zju.edu.cn

More information

Component-based Face Recognition with 3D Morphable Models

Component-based Face Recognition with 3D Morphable Models Component-based Face Recognition with 3D Morphable Models Jennifer Huang 1, Bernd Heisele 1,2, and Volker Blanz 3 1 Center for Biological and Computational Learning, M.I.T., Cambridge, MA, USA 2 Honda

More information

Better than best: matching score based face registration

Better than best: matching score based face registration Better than best: based face registration Luuk Spreeuwers University of Twente Fac. EEMCS, Signals and Systems Group Hogekamp Building, 7522 NB Enschede The Netherlands l.j.spreeuwers@ewi.utwente.nl Bas

More information

3D Signatures for Fast 3D Face Recognition

3D Signatures for Fast 3D Face Recognition 3D Signatures for Fast 3D Face Recognition Chris Boehnen, Tanya Peters, and Patrick J. Flynn Department of Computer Science and Engineering University of Notre Dame, USA {cboehnen,tpeters,flynn}@nd.edu

More information

Correspondence. CS 468 Geometry Processing Algorithms. Maks Ovsjanikov

Correspondence. CS 468 Geometry Processing Algorithms. Maks Ovsjanikov Shape Matching & Correspondence CS 468 Geometry Processing Algorithms Maks Ovsjanikov Wednesday, October 27 th 2010 Overall Goal Given two shapes, find correspondences between them. Overall Goal Given

More information

Accurate 3D Face and Body Modeling from a Single Fixed Kinect

Accurate 3D Face and Body Modeling from a Single Fixed Kinect Accurate 3D Face and Body Modeling from a Single Fixed Kinect Ruizhe Wang*, Matthias Hernandez*, Jongmoo Choi, Gérard Medioni Computer Vision Lab, IRIS University of Southern California Abstract In this

More information

3D face recognition based on a modified ICP method

3D face recognition based on a modified ICP method University of Wollongong Research Online Faculty of Informatics - Papers (Archive) Faculty of Engineering and Information Sciences 2011 3D face recognition based on a modified ICP method Kankan Zhao University

More information

A Survey of Light Source Detection Methods

A Survey of Light Source Detection Methods A Survey of Light Source Detection Methods Nathan Funk University of Alberta Mini-Project for CMPUT 603 November 30, 2003 Abstract This paper provides an overview of the most prominent techniques for light

More information

Haresh D. Chande #, Zankhana H. Shah *

Haresh D. Chande #, Zankhana H. Shah * Illumination Invariant Face Recognition System Haresh D. Chande #, Zankhana H. Shah * # Computer Engineering Department, Birla Vishvakarma Mahavidyalaya, Gujarat Technological University, India * Information

More information

Reconstruction of complete 3D object model from multi-view range images.

Reconstruction of complete 3D object model from multi-view range images. Header for SPIE use Reconstruction of complete 3D object model from multi-view range images. Yi-Ping Hung *, Chu-Song Chen, Ing-Bor Hsieh, Chiou-Shann Fuh Institute of Information Science, Academia Sinica,

More information

Occluded Facial Expression Tracking

Occluded Facial Expression Tracking Occluded Facial Expression Tracking Hugo Mercier 1, Julien Peyras 2, and Patrice Dalle 1 1 Institut de Recherche en Informatique de Toulouse 118, route de Narbonne, F-31062 Toulouse Cedex 9 2 Dipartimento

More information

Multiple Nose Region Matching for 3D Face Recognition under Varying Facial Expression

Multiple Nose Region Matching for 3D Face Recognition under Varying Facial Expression IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 28, NO. 10, OCTOBER 2006 1 Multiple Nose Region Matching for 3D Face Recognition under Varying Facial Expression Kyong I. Chang, Kevin

More information

Shape Model-Based 3D Ear Detection from Side Face Range Images

Shape Model-Based 3D Ear Detection from Side Face Range Images Shape Model-Based 3D Ear Detection from Side Face Range Images Hui Chen and Bir Bhanu Center for Research in Intelligent Systems University of California, Riverside, California 92521, USA fhchen, bhanug@vislab.ucr.edu

More information

An Algorithm to Determine the Chromaticity Under Non-uniform Illuminant

An Algorithm to Determine the Chromaticity Under Non-uniform Illuminant An Algorithm to Determine the Chromaticity Under Non-uniform Illuminant Sivalogeswaran Ratnasingam and Steve Collins Department of Engineering Science, University of Oxford, OX1 3PJ, Oxford, United Kingdom

More information

FACIAL ANIMATION FROM SEVERAL IMAGES

FACIAL ANIMATION FROM SEVERAL IMAGES International Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 5. Hakodate 1998 FACIAL ANIMATION FROM SEVERAL IMAGES Yasuhiro MUKAIGAWAt Yuichi NAKAMURA+ Yuichi OHTA+ t Department of Information

More information

Facial Animation System Design based on Image Processing DU Xueyan1, a

Facial Animation System Design based on Image Processing DU Xueyan1, a 4th International Conference on Machinery, Materials and Computing Technology (ICMMCT 206) Facial Animation System Design based on Image Processing DU Xueyan, a Foreign Language School, Wuhan Polytechnic,

More information

Scanning Real World Objects without Worries 3D Reconstruction

Scanning Real World Objects without Worries 3D Reconstruction Scanning Real World Objects without Worries 3D Reconstruction 1. Overview Feng Li 308262 Kuan Tian 308263 This document is written for the 3D reconstruction part in the course Scanning real world objects

More information

5.2 Surface Registration

5.2 Surface Registration Spring 2018 CSCI 621: Digital Geometry Processing 5.2 Surface Registration Hao Li http://cs621.hao-li.com 1 Acknowledgement Images and Slides are courtesy of Prof. Szymon Rusinkiewicz, Princeton University

More information

Linear Discriminant Analysis for 3D Face Recognition System

Linear Discriminant Analysis for 3D Face Recognition System Linear Discriminant Analysis for 3D Face Recognition System 3.1 Introduction Face recognition and verification have been at the top of the research agenda of the computer vision community in recent times.

More information

Automatic 3D Face Detection, Normalization and Recognition

Automatic 3D Face Detection, Normalization and Recognition Automatic 3D Face Detection, Normalization and Recognition Ajmal Mian, Mohammed Bennamoun and Robyn Owens School of Computer Science and Software Engineering The University of Western Australia 35 Stirling

More information

Fingerprint Indexing using Minutiae and Pore Features

Fingerprint Indexing using Minutiae and Pore Features Fingerprint Indexing using Minutiae and Pore Features R. Singh 1, M. Vatsa 1, and A. Noore 2 1 IIIT Delhi, India, {rsingh, mayank}iiitd.ac.in 2 West Virginia University, Morgantown, USA, afzel.noore@mail.wvu.edu

More information

LOCAL APPEARANCE BASED FACE RECOGNITION USING DISCRETE COSINE TRANSFORM

LOCAL APPEARANCE BASED FACE RECOGNITION USING DISCRETE COSINE TRANSFORM LOCAL APPEARANCE BASED FACE RECOGNITION USING DISCRETE COSINE TRANSFORM Hazim Kemal Ekenel, Rainer Stiefelhagen Interactive Systems Labs, University of Karlsruhe Am Fasanengarten 5, 76131, Karlsruhe, Germany

More information

Semi-Supervised PCA-based Face Recognition Using Self-Training

Semi-Supervised PCA-based Face Recognition Using Self-Training Semi-Supervised PCA-based Face Recognition Using Self-Training Fabio Roli and Gian Luca Marcialis Dept. of Electrical and Electronic Engineering, University of Cagliari Piazza d Armi, 09123 Cagliari, Italy

More information

Structural Human Shape Analysis for Modeling and Recognition

Structural Human Shape Analysis for Modeling and Recognition Structural Human Shape Analysis for Modeling and Recognition Chutisant Kerdvibulvech 1 and Koichiro Yamauchi 2 1 Rangsit University, 52/347 Muang-Ake, Paholyothin Rd, Lak-Hok, Patum Thani 12000, Thailand

More information

Detection of Facial Landmarks of North Eastern Indian People

Detection of Facial Landmarks of North Eastern Indian People International Journal of Information and Computation Technology. ISSN 0974-2239 Volume 3, Number 9 (2013), pp. 953-962 International Research Publications House http://www. irphouse.com /ijict.htm Detection

More information

Multiscale 3D Feature Extraction and Matching

Multiscale 3D Feature Extraction and Matching Multiscale 3D Feature Extraction and Matching Hadi Fadaifard and George Wolberg Graduate Center, City University of New York May 18, 2011 Hadi Fadaifard and George Wolberg Multiscale 3D Feature Extraction

More information

Illumination-Robust Face Recognition based on Gabor Feature Face Intrinsic Identity PCA Model

Illumination-Robust Face Recognition based on Gabor Feature Face Intrinsic Identity PCA Model Illumination-Robust Face Recognition based on Gabor Feature Face Intrinsic Identity PCA Model TAE IN SEOL*, SUN-TAE CHUNG*, SUNHO KI**, SEONGWON CHO**, YUN-KWANG HONG*** *School of Electronic Engineering

More information

Structured light 3D reconstruction

Structured light 3D reconstruction Structured light 3D reconstruction Reconstruction pipeline and industrial applications rodola@dsi.unive.it 11/05/2010 3D Reconstruction 3D reconstruction is the process of capturing the shape and appearance

More information

Classification of Face Images for Gender, Age, Facial Expression, and Identity 1

Classification of Face Images for Gender, Age, Facial Expression, and Identity 1 Proc. Int. Conf. on Artificial Neural Networks (ICANN 05), Warsaw, LNCS 3696, vol. I, pp. 569-574, Springer Verlag 2005 Classification of Face Images for Gender, Age, Facial Expression, and Identity 1

More information

Image Processing Pipeline for Facial Expression Recognition under Variable Lighting

Image Processing Pipeline for Facial Expression Recognition under Variable Lighting Image Processing Pipeline for Facial Expression Recognition under Variable Lighting Ralph Ma, Amr Mohamed ralphma@stanford.edu, amr1@stanford.edu Abstract Much research has been done in the field of automated

More information

IDENTITY VERIFICATION VIA THE 3DID FACE ALIGNMENT SYSTEM. Dirk Colbry and George Stockman

IDENTITY VERIFICATION VIA THE 3DID FACE ALIGNMENT SYSTEM. Dirk Colbry and George Stockman IDENTITY VERIFICATION VIA THE 3DID FACE ALIGNMENT SYSTEM Dirk Colbry and George Stockman Department of Computer Science and Engineering, Michigan State University East Lansing, Michigan 48824-1226 {colbrydi,

More information

Introduction to Mobile Robotics Iterative Closest Point Algorithm. Wolfram Burgard, Cyrill Stachniss, Maren Bennewitz, Kai Arras

Introduction to Mobile Robotics Iterative Closest Point Algorithm. Wolfram Burgard, Cyrill Stachniss, Maren Bennewitz, Kai Arras Introduction to Mobile Robotics Iterative Closest Point Algorithm Wolfram Burgard, Cyrill Stachniss, Maren Bennewitz, Kai Arras 1 Motivation 2 The Problem Given: two corresponding point sets: Wanted: translation

More information

A Study on the Consistency of Features for On-line Signature Verification

A Study on the Consistency of Features for On-line Signature Verification A Study on the Consistency of Features for On-line Signature Verification Center for Unified Biometrics and Sensors State University of New York at Buffalo Amherst, NY 14260 {hlei,govind}@cse.buffalo.edu

More information

Expression-Invariant 3D Face Recognition using Patched Geodesic Texture Transform

Expression-Invariant 3D Face Recognition using Patched Geodesic Texture Transform 2010 Digital Image Computing: Techniques and Applications Expression-Invariant 3D Face Recognition using Patched Geodesic Texture Transform Farshid Hajati 1, 2, Abolghasem A. Raie 1 1 Faculty of Electrical

More information

3D Digitization of Human Foot Based on Computer Stereo Vision Combined with KINECT Sensor Hai-Qing YANG a,*, Li HE b, Geng-Xin GUO c and Yong-Jun XU d

3D Digitization of Human Foot Based on Computer Stereo Vision Combined with KINECT Sensor Hai-Qing YANG a,*, Li HE b, Geng-Xin GUO c and Yong-Jun XU d 2017 International Conference on Mechanical Engineering and Control Automation (ICMECA 2017) ISBN: 978-1-60595-449-3 3D Digitization of Human Foot Based on Computer Stereo Vision Combined with KINECT Sensor

More information

Robust Facial Feature Detection for Registration

Robust Facial Feature Detection for Registration Robust Facial Feature Detection for Registration Taher KHADHRAOUI khadhra.th@gmail.com Fouazi BENZARTI benzartif@yahoo.fr Hamid AMIRI hamidlamiri@yahoo.com Abstract Face image analysis, in the context

More information

TEXTURE OVERLAY ONTO NON-RIGID SURFACE USING COMMODITY DEPTH CAMERA

TEXTURE OVERLAY ONTO NON-RIGID SURFACE USING COMMODITY DEPTH CAMERA TEXTURE OVERLAY ONTO NON-RIGID SURFACE USING COMMODITY DEPTH CAMERA Tomoki Hayashi 1, Francois de Sorbier 1 and Hideo Saito 1 1 Graduate School of Science and Technology, Keio University, 3-14-1 Hiyoshi,

More information

Algorithms for 3D Isometric Shape Correspondence

Algorithms for 3D Isometric Shape Correspondence Algorithms for 3D Isometric Shape Correspondence Yusuf Sahillioğlu Computer Eng. Dept., Koç University, Istanbul, Turkey (PhD) Computer Eng. Dept., METU, Ankara, Turkey (Asst. Prof.) 2 / 53 Problem Definition

More information

Fast and Accurate 3D Face Recognition

Fast and Accurate 3D Face Recognition Int J Comput Vis (2011) 93: 389 414 DOI 10.1007/s11263-011-0426-2 Fast and Accurate 3D Face Recognition Using Registration to an Intrinsic Coordinate System and Fusion of Multiple Region Classifiers Luuk

More information

Training Algorithms for Robust Face Recognition using a Template-matching Approach

Training Algorithms for Robust Face Recognition using a Template-matching Approach Training Algorithms for Robust Face Recognition using a Template-matching Approach Xiaoyan Mu, Mehmet Artiklar, Metin Artiklar, and Mohamad H. Hassoun Department of Electrical and Computer Engineering

More information

Intensity-Depth Face Alignment Using Cascade Shape Regression

Intensity-Depth Face Alignment Using Cascade Shape Regression Intensity-Depth Face Alignment Using Cascade Shape Regression Yang Cao 1 and Bao-Liang Lu 1,2 1 Center for Brain-like Computing and Machine Intelligence Department of Computer Science and Engineering Shanghai

More information

Face Recognition Technology Based On Image Processing Chen Xin, Yajuan Li, Zhimin Tian

Face Recognition Technology Based On Image Processing Chen Xin, Yajuan Li, Zhimin Tian 4th International Conference on Machinery, Materials and Computing Technology (ICMMCT 2016) Face Recognition Technology Based On Image Processing Chen Xin, Yajuan Li, Zhimin Tian Hebei Engineering and

More information

Global fitting of a facial model to facial features for model based video coding

Global fitting of a facial model to facial features for model based video coding Global fitting of a facial model to facial features for model based video coding P M Hillman J M Hannah P M Grant University of Edinburgh School of Engineering and Electronics Sanderson Building, King

More information

EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm

EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm EE368 Project Report CD Cover Recognition Using Modified SIFT Algorithm Group 1: Mina A. Makar Stanford University mamakar@stanford.edu Abstract In this report, we investigate the application of the Scale-Invariant

More information

An Adaptive Threshold LBP Algorithm for Face Recognition

An Adaptive Threshold LBP Algorithm for Face Recognition An Adaptive Threshold LBP Algorithm for Face Recognition Xiaoping Jiang 1, Chuyu Guo 1,*, Hua Zhang 1, and Chenghua Li 1 1 College of Electronics and Information Engineering, Hubei Key Laboratory of Intelligent

More information

Expression Invariant 3D Face Recognition with a Morphable Model

Expression Invariant 3D Face Recognition with a Morphable Model Expression Invariant 3D Face Recognition with a Morphable Model Brian Amberg brian.amberg@unibas.ch Reinhard Knothe reinhard.knothe@unibas.ch Thomas Vetter thomas.vetter@unibas.ch Abstract We describe

More information

Object. Radiance. Viewpoint v

Object. Radiance. Viewpoint v Fisher Light-Fields for Face Recognition Across Pose and Illumination Ralph Gross, Iain Matthews, and Simon Baker The Robotics Institute, Carnegie Mellon University 5000 Forbes Avenue, Pittsburgh, PA 15213

More information

Facial Expression Detection Using Implemented (PCA) Algorithm

Facial Expression Detection Using Implemented (PCA) Algorithm Facial Expression Detection Using Implemented (PCA) Algorithm Dileep Gautam (M.Tech Cse) Iftm University Moradabad Up India Abstract: Facial expression plays very important role in the communication with

More information

Graph Matching Iris Image Blocks with Local Binary Pattern

Graph Matching Iris Image Blocks with Local Binary Pattern Graph Matching Iris Image Blocs with Local Binary Pattern Zhenan Sun, Tieniu Tan, and Xianchao Qiu Center for Biometrics and Security Research, National Laboratory of Pattern Recognition, Institute of

More information

Geometric Modeling and Processing

Geometric Modeling and Processing Geometric Modeling and Processing Tutorial of 3DIM&PVT 2011 (Hangzhou, China) May 16, 2011 4. Geometric Registration 4.1 Rigid Registration Range Scanning: Reconstruction Set of raw scans Reconstructed

More information

Iterative Closest Point Algorithm in the Presence of Anisotropic Noise

Iterative Closest Point Algorithm in the Presence of Anisotropic Noise Iterative Closest Point Algorithm in the Presence of Anisotropic Noise L. Maier-Hein, T. R. dos Santos, A. M. Franz, H.-P. Meinzer German Cancer Research Center, Div. of Medical and Biological Informatics

More information

Abstract We present a system which automatically generates a 3D face model from a single frontal image of a face. Our system consists of two component

Abstract We present a system which automatically generates a 3D face model from a single frontal image of a face. Our system consists of two component A Fully Automatic System To Model Faces From a Single Image Zicheng Liu Microsoft Research August 2003 Technical Report MSR-TR-2003-55 Microsoft Research Microsoft Corporation One Microsoft Way Redmond,

More information

Scan Matching. Pieter Abbeel UC Berkeley EECS. Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics

Scan Matching. Pieter Abbeel UC Berkeley EECS. Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics Scan Matching Pieter Abbeel UC Berkeley EECS Many slides adapted from Thrun, Burgard and Fox, Probabilistic Robotics Scan Matching Overview Problem statement: Given a scan and a map, or a scan and a scan,

More information

Fusion of Infrared and Range Data: Multi-modal Face Images

Fusion of Infrared and Range Data: Multi-modal Face Images Fusion of Infrared and Range Data: Multi-modal Face Images Xin Chen, Patrick J. Flynn, and Kevin W. Bowyer Dept. of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN 46556 USA

More information

DEFORMABLE MATCHING OF HAND SHAPES FOR USER VERIFICATION. Ani1 K. Jain and Nicolae Duta

DEFORMABLE MATCHING OF HAND SHAPES FOR USER VERIFICATION. Ani1 K. Jain and Nicolae Duta DEFORMABLE MATCHING OF HAND SHAPES FOR USER VERIFICATION Ani1 K. Jain and Nicolae Duta Department of Computer Science and Engineering Michigan State University, East Lansing, MI 48824-1026, USA E-mail:

More information

Free-form 3D object reconstruction from range images. C. Schütz, T. Jost, H. Hügli

Free-form 3D object reconstruction from range images. C. Schütz, T. Jost, H. Hügli Free-form 3D object reconstruction from range images C. Schütz, T. Jost, H. Hügli Institute for Microtechnology University of Neuchatel, rue Breguet 2 CH-2000 Neuchatel, Switzerland email: christian.schutz@imt.unine.ch

More information

Using Augmented Measurements to Improve the Convergence of ICP. Jacopo Serafin and Giorgio Grisetti

Using Augmented Measurements to Improve the Convergence of ICP. Jacopo Serafin and Giorgio Grisetti Jacopo Serafin and Giorgio Grisetti Point Cloud Registration We want to find the rotation and the translation that maximize the overlap between two point clouds Page 2 Point Cloud Registration We want

More information

Perception and Action using Multilinear Forms

Perception and Action using Multilinear Forms Perception and Action using Multilinear Forms Anders Heyden, Gunnar Sparr, Kalle Åström Dept of Mathematics, Lund University Box 118, S-221 00 Lund, Sweden email: {heyden,gunnar,kalle}@maths.lth.se Abstract

More information

Rotation Invariant Finger Vein Recognition *

Rotation Invariant Finger Vein Recognition * Rotation Invariant Finger Vein Recognition * Shaohua Pang, Yilong Yin **, Gongping Yang, and Yanan Li School of Computer Science and Technology, Shandong University, Jinan, China pangshaohua11271987@126.com,

More information

Integrated 3D Expression Recognition and Face Recognition

Integrated 3D Expression Recognition and Face Recognition Integrated 3D Expression Recognition and Face Recognition Chao Li, Armando Barreto Electrical & Computer Engineering Department Florida International University Miami, Florida, 33174, USA {cli007, barretoa}@fiu.edu

More information

Experiments with Edge Detection using One-dimensional Surface Fitting

Experiments with Edge Detection using One-dimensional Surface Fitting Experiments with Edge Detection using One-dimensional Surface Fitting Gabor Terei, Jorge Luis Nunes e Silva Brito The Ohio State University, Department of Geodetic Science and Surveying 1958 Neil Avenue,

More information