Automatic Class Selection and Prototyping for 3-D Object Classification

Abstract

Most research on 3-D object classification and recognition focuses on recognition of objects in 3-

Three-Dimensional Model-Based Object Recognition and Segmentation in Cluttered Scenes

“…Next, the 2D distances between the model vertices and their respective nearest scene vertices are calculated. These distances are sorted and the longest n 1 distances are used to calculate the active space violation measure using (12).…”

“…Recently, the spin image representation was used in a batch RANSAC algorithm for rapid 3D vehicle recognition [37]. It was also used for automatic clustering and classification of 3D vehicles [12], [21] in order to handle large databases and classify unknown but similar vehicles. Some pairwise correspondence techniques (e.g., [1], [8]) have also been applied to 3D object recognition.…”

Iterative 3D shape classification by online metric learning

“…Most existing shape classification methods usually use the labeled model set as the training data, and train a classifier based on the supervised learning methods, such as nearest neighbor classifier (Csakany and Wallace, 2003;Donamukkala et al, 2005;Biasotti et al, 2006), Bayesian classifier (Huber et al, 2004), SVM (Marini et al, 2011;Barra and Biasotti, 2014), belief function (Tabia et al, 2013) and deep neural network classifier (Qin et al, 2014). A recent work presented by Huang et al (2013b) uses a semi-supervised method for fine-grained 3D shape classification with several pre-labeled samples.…”

“…Firstly, no pre-labeled samples and pre-trained classifiers are required. The unsupervised clustering method (Everitt et al, 2011) is intuitively one of the best choices as used in the 3D shape classification (Donamukkala et al, 2005) and co-segmentation (Wang et al, 2012). Unfortunately, the class number of the data sets, which affects the clustering result heavily, must be the default known or specified by the user for most of the clustering methods, and the data sets are often classified in a one-pass batch process without any other user intervention.…”

Non-rigid 3D shape classification using bag-of-feature techniques

“…Organizing these libraries into categories and providing effective indexing is imperative for real time browsing and retrieval. Some paper such as [1,2] have addressed the categorization problem. Their methods distinguish the training step, in which the classes of the database are constructed, from the actual classification, that associates the object model to one class.…”