Halim Benhabiles

Learning Boundary Edges for 3D-Mesh Segmentation

This paper presents a 3D‐mesh segmentation algorithm based on a learning approach. A large database of manually segmented 3D‐meshes is used to learn a boundary edge function. The function is learned using a classifier which automatically selects from a pool of geometric features the most relevant ones to detect candidate boundary edges. We propose a processing pipeline that produces smooth closed boundaries using this edge function. This pipeline successively selects a set of candidate boundary contours, closes them and optimizes them using a snake movement. Our algorithm was evaluated quantitatively using two different segmentation benchmarks and was shown to outperform most recent algorithms from the state‐of‐the‐art.

A framework for the objective evaluation of segmentation algorithms using a ground-truth of human segmented 3D-models

In this paper, we present an evaluation method of 3D-mesh segmentation algorithms based on a ground-truth corpus. This corpus is composed of a set of 3D-models grouped in different classes (animals, furnitures, etc.) associated with several manual segmentations produced by human observers. We define a measure that quantifies the consistency between two segmentations of a 3D-model, whatever their granularity. Finally, we propose an objective quality score for the automatic evaluation of 3D-mesh segmentation algorithms based on these measures and on the ground-truth corpus. Thus the quality of segmentations obtained by automatic algorithms is evaluated in a quantitative way thanks to the quality score, and on an objective basis thanks to the groundtruth corpus. Our approach is illustrated through the evaluation of two recent 3D-mesh segmentation methods.

A comparative study of existing metrics for 3D-mesh segmentation evaluation

In this paper, we present an extensive experimental comparison of existing similarity metrics addressing the quality assessment problem of mesh segmentation. We introduce a new metric, named the 3D Normalized Probabilistic Rand Index (3D-NPRI), which outperforms the others in terms of properties and discriminative power. This comparative study includes a subjective experiment with human observers and is based on a corpus of manually segmented models. This corpus is an improved version of our previous one (Benhabiles et al. in IEEE International Conference on Shape Modeling and Application (SMI), 2009). It is composed of a set of 3D-mesh models grouped in different classes associated with several manual ground-truth segmentations. Finally the 3D-NPRI is applied to evaluate six recent segmentation algorithms using our corpus and the Chen et al.’s (ACM Trans. Graph. (SIGGRAPH), 28(3), 2009) corpus.

SHREC'12 track: 3D mesh segmentation

3D mesh segmentation is a fundamental process in many applications such as shape retrieval, compression, deformation, etc. The objective of this track is to evaluate the performance of recent segmentation methods using a ground-truth corpus and an accurate similarity metric. The ground-truth corpus is composed of 28 watertight models, grouped in five classes (animal, furniture, hand, human and bust) and each associated with 4 ground-truth segmentations done by human subjects. 3 research groups have participated to this track, the accuracy of their segmentation algorithms have been evaluated and compared with 4 other state-of-the-art methods.

Fast simplification with sharp feature preserving for 3D point clouds

This paper presents a fast point cloud simplification method that allows to preserve sharp edge points. The method is based on the combination of both clustering and coarse-to-fine simplification approaches. It consists to firstly create a coarse cloud using a clustering algorithm. Then each point of the resulting coarse cloud is assigned a weight that quantifies its importance, and allows to classify it into a sharp point or a simple point. Finally, both kinds of points are used to refine the coarse cloud and thus create a new simplified cloud characterized by high density of points in sharp regions and low density in flat regions. Experiments show that our algorithm is much faster than the last proposed simplification algorithm [1] which deals with sharp edge points preserving, and still produces similar results.

Kinematic skeleton extraction based on motion boundaries for 3D dynamic meshes

This paper presents a precise kinematic skeleton extraction method for 3D dynamic meshes. Contrary to previous methods, our method is based on the computation of motion boundaries instead of detecting object parts characterized by rigid transformations. Thanks to a learned boundary edge function, we are able to compute efficiently a set of motion boundaries which in fact correspond to all possible articulations of the 3D object. Moreover, the boundaries are detected even if the parts linked to an objects articulation are immobile over time. The different boundaries are then used to extract the kinematic skeleton. Experiments show that our algorithm produces more precise skeletons compared to previous methods.

Convolutional neural network for pottery retrieval

Abstract. The effectiveness of the convolutional neural network (CNN) has already been demonstrated in many challenging tasks of computer vision, such as image retrieval, action recognition, and object classification. This paper specifically exploits CNN to design local descriptors for content-based retrieval of complete or nearly complete three-dimensional (3-D) vessel replicas. Based on vector quantization, the designed descriptors are clustered to form a shape vocabulary. Then, each 3-D object is associated to a set of clusters (words) in that vocabulary. Finally, a weighted vector counting the occurrences of every word is computed. The reported experimental results on the 3-D pottery benchmark show the superior performance of the proposed method.

Belief-Function-Based Framework for Deformable 3D-Shape Retrieval

The need for efficient tools to index and retrieve 3D content becomes even more acute. This paper presents a fully automatic 3D-object retrieval method. It consists of two main steps namely shape signature extraction to describe the shape of objects, and similarity computing to compute similarity between objects. In the first step (signature extraction), we use a shape descriptor called geodesic cords. This descriptor can be seen as a probability distribution sampled from a shape function. In the second step (similarity computing), a global distance, based on belief function theory, is computed between each pair wise of descriptors corresponding respectively to an object query and an object from a given database. Experiments on commonly-used benchmarks demonstrate that our method obtains competitive performance compared to 3D-object retrieval methods from the state-of-the-art.

Dynamic 3D Facial Expression Recognition Using Robust Shape Features

In this paper we present a novel approach for dynamic facial expression recognition based on 3D geometric facial features. Geodesic distances between corresponding 3D open curves are computed and used as features to describe the facial changes across sequences of 3D face scans. Hidden Markov Models (HMMs) are exploited to learn the curves shape variation through a 3D frame sequences, and the trained models are used to classify six prototypic facial expressions. Our approach shows high performance, and an overall recognition rate of 94.45% is attained after a validation on the BU-4DFE database.