3D face landmark labelling

Creusot, Clement; Pears, Nick; Austin, Jim

doi:10.1145/1877808.1877815

Cited by 28 publications

(25 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As expected, the deformable model is superior to the rigid one and this is verified for all tested landmarks. Since variations in our dataset are mainly due to identity with only residual expression changes, this is a remarkable result in favor of non-rigid modeling vis-à-vis the rigid registration used, for example, by Creusot et al [12] and Ambert & Vetter [13]. The median errors of SRILF were below 3 mm for all landmarks except the chin tip and outer eye corners.…”

Section: Localization Accuracymentioning

confidence: 70%

“…To alleviate this problem, Creusot et al [12] use partial graph matching and determine the final alignment by clustering transformations from triplets of points while Amberg & Vetter [13] use Branch and Bound to optimize the search of extended sets of landmarks (so that the missing ones are less important). However, in both cases a rigid shape is used, which is an important limitation for facial modeling.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

3D Facial Landmark Localization Using Combinatorial Search and Shape Regression

Sukno

Waddington

Whelan

2012

Computer Vision – ECCV 2012. Workshops and Demonstrations

View full text Add to dashboard Cite

Abstract. This paper presents a method for the automatic detection of facial landmarks. The algorithm receives a set of 3D candidate points for each landmark (e.g. from a feature detector) and performs combinatorial search constrained by a deformable shape model. A key assumption of our approach is that for some landmarks there might not be an accurate candidate in the input set. This is tackled by detecting partial subsets of landmarks and inferring those that are missing so that the probability of the deformable model is maximized. The ability of the model to work with incomplete information makes it possible to limit the number of candidates that need to be retained, substantially reducing the number of possible combinations to be tested with respect to the alternative of trying to always detect the complete set of landmarks. We demonstrate the accuracy of the proposed method in a set of 144 facial scans acquired by means of a hand-held laser scanner in the context of clinical craniofacial dysmorphology research. Using spin images to describe the geometry and targeting 11 facial landmarks, we obtain an average error below 3 mm, which compares favorably with other state of the art approaches based on geometric descriptors.

show abstract

Section: Localization Accuracymentioning

confidence: 70%

Section: Related Workmentioning

confidence: 99%

3D Facial Landmark Localization Using Combinatorial Search and Shape Regression

Sukno

Waddington

Whelan

2012

Computer Vision – ECCV 2012. Workshops and Demonstrations

View full text Add to dashboard Cite

show abstract

“…An aspect of our system is that it cannot discriminate between shapes of interest that are linked to identity from the ones linked to change in expression or other variations. We think that the best way to use those detected points will be to label them (for example with [2]) to get a more sparse and consistent set of points. …”

Section: Resultsmentioning

confidence: 99%

“…Finding correspondences between 3D surfaces is a crucial step in many 3D shape processing applications, such as surface landmarking [2], surface registration [3], 3D object retrieval [4] and 3D face recognition [5]. The difficulty in solving this correspondence problem is dependent on the type of shapes that need to be matched.…”

Section: Introductionmentioning

confidence: 99%

Automatic Keypoint Detection on 3D Faces Using a Dictionary of Local Shapes

Creusot

Pears

Austin

2011

2011 International Conference on 3D Imaging, Modeling, Processing, Visualization and Transmission

Self Cite

View full text Add to dashboard Cite

Abstract-Keypoints on 3D surfaces are points that can be extracted repeatably over a wide range of 3D imaging conditions. They are used in many 3D shape processing applications; for example, to establish a set of initial correspondences across a pair of surfaces to be matched. Typically, keypoints are extracted using extremal values of a function over the 3D surface, such as the descriptor map for Gaussian curvature. That approach works well for salient points, such as the nosetip, but can not be used with other less pronounced local shapes. In this paper, we present an automatic method to detect keypoints on 3D faces, where these keypoints are locally similar to a set of previously learnt shapes, constituting a 'local shape dictionary'. The local shapes are learnt at a set of 14 manuallyplaced landmark positions on the human face. Local shapes are characterised by a set of 10 shape descriptors computed over a range of scales. For each landmark, the proportion of face meshes that have an associated keypoint detection is used as a performance indicator. Repeatability of the extracted keypoints is measured across the FRGC v2 database.

show abstract

“…For instance, it is well known that even holistic matching methods, such as Eigenfaces [8] and Fisherfaces [9], need accurate locations of key facial features for face pose normalisation; where noticeable degradation in recognition performance is observed without accurate facial feature locations. Furthermore, it is generally believed that, an improved landmark localisation will increase the effectiveness of many face processing applications [11][12][13][14][15].…”

Section: Introductionmentioning

confidence: 99%