A Comparative Survey on 3D Models Retrieval Methods

Abstract

Abstract-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 purpose two main approaches used in the retrieval of 3D deformable models will be reviewed and implemented: the spectral approach and the bag-of-features approach. The deformation or variation in shape involves different aspects depending on the type of object. It could be a change in the posture of an articulated or bendable model, or it could result from a variation (loss or gain) in the total mass leading to a change in the surface and thus in the shape of the object. Even more complex situations occur when both cases are combined together.

3D Model Identification Using Weighted Implicit Shape Representation and Panoramic View

“…Several researches have been conducted to group 3D models into corresponding categories by matching the features and comparing the similarities of the models [5][6][7][8][9][10][11]14,[19][20][21][22]. The matching and comparison are implemented on a huge dataset containing various models of different poses and shapes.…”

“…3D model identification methods are classified into two categories: view-based and model-based methods [6,7]. Model based methods include geometric moment [19], volumetric descriptor [20], surface distribution [21], surface geometry [22].…”

“…However, the geometry and topology based methods are generally computationally cost and are fragile to 3D model removal. View-based methods have a high discriminative property for 3D model representation [5][6][7][8][9][10][11]. A 2D view image is a range image obtained from a viewpoint located on a 3D model's bounding sphere.…”