3D object indexing and recognition

Image and Vision Computing, 2001

The recognition of free-form 3D objects using 3D models under different viewing conditions based on the geometric hashing algorithm and global veri®cation is presented. The matching stage of the algorithm uses the hash-table prepared in the off-line stage. Given a scene of feature points, one tries to match the measurements taken at scene points to those memorised in the hash-table. The technique used for feature recovery is the generalisation of the CSS method (IEEE Trans. Pattern Anal. Mach. Intell., 14 (1992) 789±805), which is a powerful shape descriptor expected to be an MPEG-7 standard. Smoothing is used to remove noise and reduce the number of feature points to add to the ef®ciency and robustness of the system. The local maxima of Gaussian and mean curvatures are selected as feature points. Furthermore, the torsion maxima of the zero-crossing contours of Gaussian and mean curvatures are also selected as feature points. Recognition results are demonstrated for rotated and scaled as well as partially occluded objects. In order to verify match, 3D translation, rotation and scaling parameters are used for veri®cation and results indicate that our technique is invariant to those transformations. Our technique for smoothing and feature extraction is more suitable than level set methods or volumetric diffusion for object recognition applications since it is applicable to incomplete surface data that arise during occlusion. It is also more ef®cient and allows for accurate estimation of curvature values. q

Image and Vision Computing, 1992

SPIE Proceedings, 2010

This paper proposes a comprehensive overview of state of the art 2D/3D, view-based indexing methods. The principle of 2D/3D indexing methods consists of describing 3D models by means of a set of 2D shape descriptors, associated with a set of corresponding 2D views (under the assumption of a given projection model). Notably, such an approach makes it possible to identify 3D objects of interest from 2D images/videos. An experimental evaluation is also proposed, in order to examine the influence of the number of views and of the associated viewing angle selection strategies on the retrieval results. Experiments concern both 3D model retrieval and image recognition from a single view. Results obtained show promising performances, with recognition rates from a single view higher then 66%, which opens interesting perspectives in terms of semantic metadata extraction from still images/videos.

Journal of Electronic Commerce in Organizations, 2015

This paper focuses on the recognition of 3D objects using 2D attributes. In order to increase the recognition rate, the present an hybridization of three approaches to calculate the attributes of color image, this hybridization based on the combination of Zernike moments, Gist descriptors and color descriptor (statistical moments). In the classification phase, three methods are adopted: Neural Network (NN), Support Vector Machine (SVM), and k-nearest neighbor (KNN). The database COIL-100 is used in the experimental results.

This paper describes 3D object searching mechanism and methods which mainly focuses on query interfaces and their corresponding model/query representations, matching methods and algorithms. Many 3D models are available online but finding the right tool requires great effort, time and resources to develop a high quality 3D model. However finding the right model is not easy as they are scattered across the web on different repository sites, project sites, and personal homepages. It presents a new approach to 3Dimensional object recognition by using an Octree model library (OML) I, II and fast search algorithm [1]. The fast search algorithm is used for locating four pairs of feature points to estimate the viewing direction on effective 2 level database. The best direction is decided by searching for the four pairs of feature points between the input image and image projected along the approximate viewing direction. The input shape is found by matching it to the projected shape. The complexity of proposed method is shown to be O(n 2) in the worst case, and that of the simple combinatorial method of O(m4 , n2), where n and m denote the number of feature points of the 3D model object and the 2D object, respectively.

2000

This paper addresses the problem of recognizing 3D objects from 2D intensity images. It describes the object recognition system named RIO (relational indexing of objects), which contains a number of new techniques. RIO begins with an edge image obtained from a pair of intensity images taken with a single camera and two different lightings.

We propose techniques for designing and training of pose-invariant object recognition systems using realistic 3d computer graphics models. We look at the relation between the size of the training set and the classification accuracy for a basic recognition task and provide a method for estimating the degree of difficulty of detecting an object. We show how to sample, align, and cluster images of objects on the view sphere. We address the problem of training on large, highly redundant data and propose a novel active learning method which generates compact training sets and compact classifiers.

REV Journal on Electronics and Communications, 2013

In computer vision many studies have been conducted in order to perform the matching and comparison of 3D models of objects. The main goal of matching is to group the models into different categories according to their similarity in order to allow their retrieval for recognition purposes and for further usage. So, in most of the cases, the comparison is run on a large dataset containing various models whether they belong to the same type of object or not and generally having similar or different shapes and poses. The objects’ nature and characteristics are important factors to be taken into consideration before performing the comparison step. We distinguish between two main categories of objects: rigid objects and deformable objects whose treatment and handling differ in the modeling as well as in the comparison phases. In this paper, we will be focusing on the comparison of deformable objects, and thus dealing with objects whose shapes might vary in different instances. For this pu...

Computer Vision—ECCV'96, 1996

We address the problem of 3D object recognition from a single 2D image using a model database. We develop a new method called enhanced geometric hashing. This approach allows us to solve for the indexing and the matching problem in one pass with linear complexity. Use of quasi-invariants allows us to index images in a new type of geometric hashing table. They include topological information of the observed objects inducing a high numerical stability.

Lecture Notes in Computer Science, 1990

The Geometric Hashing paradigm for model-based recognition of objects in cluttered scenes is dis~ cussed. This paradigm enables a unified approach to rigid object recognition under different viewing transformation assumptions both for 2-D and 3-D objects obtained by different sensors, e.g. vision, range, tactile. It is based on an intensive off-line model preprocessing (learning) stage, where model information is indexed into a hash-table using minimal, transformation invariant features. This enables the on-line recognition algorithm to be particularly efficient. The algorithm is straightforwardly parallelizable. Initial experimentation of the technique has led to successful recognition of both 2-D and 3-D objects in cluttered scenes from an arbitrary viewpoint. We, also, compare the Geometric tIashing with the Hough Transform and the alignment techniques. Extensions of the basic paradigm which reduce its worst case recognition complexity are discussed.